Tableau Full Course – Learn Tableau in 6 Hours | Tableau Training for Beginners | Edureka

Tableau is not just a BI tool. It is a way to tell a story to your company stakeholders based on your company data. Hi all I welcome you to this full course session on Tableau and what follows is going to be a tableau training for beginners which covers all the concepts that you need to start out with this technology. But before we begin, let's look at our agenda for today, so Going to start out with some Basics. We're going to talk about data visualization and Tableau. And why do we need both of these followed by which we are going to discuss installation of the Tableau desktop tool. Now, this is the tool that you are going to be using throughout the course of this video. Next. We are going to talk about data visualization using Tableau and then we are going to discuss an important concept called visual perception. We're also going to talk about how Tableau incorporates these visual Perceptions in Its components to give you Optimum data visualization.

Then we're going to talk about charts and graphs in Tableau and how you can enhance your data using these then we're going to incorporate it into one single dashboard using the Tableau desktop app. Now up to this point all the topics that we're going to talk about are going to be majorly focused on data visualization. Now this point onwards the topics we are going to discuss are going to be more about data aggregation. So first stop in that is Tableau functions. Now, this is tableau version of Dax next we're going to discuss level of detail another very important very fundamental concept. Then we are going to discuss parameters and Tableau followed by which we are going to talk about data blending in Tableau and how it is different from SQL joins moving on. We're going to talk about The Career aspect in Tableau. So we have a module called how to become a tableau developer which discusses the roles and responsibilities of a tableau developer.

To cap of the session, We have some interview questions for you based on Tableau. Now your instructors for this particular session is going to be Reshma and Upasana. They're going to take you through step by step through all the sessions of this Tableau tutorial. Also go ahead and kindly hit the Subscribe button. So you never miss an update from the Edureka YouTube channel. Hey, everyone. this is Reshma from Edureka. And today we'll be learning about Tableau. So let us first understand what is data visualization. Well data visualization means representing your data in a pictorial form. It may be in the form of a graph or a bar diagrams or a different kind of charts and visualization allows us to visual access to huge amounts of data in easily digestible visuals because let's say you might have got all the data and Excel sheets. You have got all that with you.

You have got the text. You've got the numbers and everything. But if you just view only numbers and text, you might not get a whole picture out of it. So you need to represent in a manner so that you can understand it better and visualization enables you to have a well-defined overview of your entire data and the simpler your visualization is the more insights and inferences you can make from it. So simple representations are the most As powerful ones so that is why we need data visualization to understand our data in a better way and these can be used for analysis of data to make future predictions. And this is highly used in solving business problems. And the two that we are going to learn today, which is Tableau This is highly used in bi so this is data visualization. Now, let's take an example. So here we have got the X and Y coordinates of different point and these represent. Resent a line. So the example that is before you this represents the data points in four quadrants. You have got the X and Y coordinates.

So if you see this data, you'll see that there is not much differences in the numbers and you might think that when you plot this data points, this might look the same but now let us take a look. So when you picture I said you'll see how different they are from each other. They are not similar at all. Even though the numbers look similar so that is why you need to visualize. It to understand it. You will not get the whole picture when you are seeing just numbers but when you plot it, you can see how different they are. So that is why data visualization is important and that is why it is highly used because picture Rising your data and analyzing will be so much easier when you plot it and you see the behavior of your data and that is how you can make future predictions.

And that is why data visualization is so popular and people have been using it all around the world. So I hope you have understood the importance of data visualization from this example. So let us move forward to our next topic and let us see the scope of visual analysis X so visual analytics is used widely. It can be used for informational it except for geospatial analysis scientific analysis takes knowledge Discovery data management and knowledge representation for presentation production and dissemination this used for a cognitive and perceptual science.

For interaction and there are many more usage of it. And why do we use visual analysis X because it helps us to make better decisions because when you can study the behavior of it, you can make better analysis of the data and you can take decisions which will be beneficial for your company and you can make future predictions accordingly and plan everything and you can also get a better sense of risk because when you can make future predictions correctly, obviously the risk factor, Go down, and this is very much beneficial for your company and you can also build better customer relationships better key stratigic initiative and better financial performance because the risk factor will be low. You can save a lot of money by studying the data by representing it and it will give you a brief idea of everything.

So this is the entire scope of visual analysis. Now, let us understand how does data visualization actually works. So the first thing you need to visualize the data, Data is a data set. Now your data set can be in form of text file. It can be any kind of flat file Excel sheets. You can also connect to any server or any database. So and not just one you can integrate and connect different data sets together. And then you analyze that data according to the parameters and then you carry out the visualization of how you want to represent what you have analyzed.

This is a brief overview of how data visualization actually works so there. Analyzing you use different formulas use different algorithms to analyze it and for visualization, you can choose different charts or Maps graphs or anything that you want whichever fits for your data set and now let us understand why Tableau if you want to visualize your data, why go for Tableau, let us understand it by looking at the features of tableau. So the first feature is that it is very flexible. You can connect it to any kind of data. It consists of amount of optimize data connectors for databases. You can connect it to an Excel file. You can connect it to a text file even collected to a Json file or any kind of server. You can connect it to a tableau server a Microsoft SQL Server Oracle Amazon redshift and many more and it provides you with a very intuitive platform.

Now according to Gartner. Tableau is actually Considered the gold standard for intuitive interactive visual analysis and an established Enterprise platform and you can represent your data in any way that you want the visuals that tableau gives you a very interactive you can tweak them around you can play around those graphs and the different charts that you make in Tableau and you can visualize your data in many ways. And also it has very quick production them. It takes only a few seconds for Tableau to wait the visualization that you want for your data. And that is the reason why Tableau has been among the top charts when it comes to a visual analogy takes and a presentation tool and let me tell you that and Tableau has managed to be in the Gartner's magic quadrant from years from now and now it is treated as the top interactive tool that is used in bi standing above click View and other tools and that is why we should all start using Tableau and it is very fun to play around with Tableau.

Let me tell you. Show you that in a while how we can make different analysis of data by representing it visually, so now let us move on to our next topic and let us see in detail. What is Tableau? So Tableau is a software company which produces interactive data visualization products focused on business intelligence and a lot of companies and almost all the big giants are using Tableau for business intelligence and for data analysis purpose.

So with Tableau, you can create and distribute interactive and shareable dashboards, which depict the trends variations and density of data informs of graphs and charts and these software also enables you data blending real-time collapse, which is what makes Tableau stand out from all its competitors and it is very unique you can use it for business purposes for academic purpose or for any purpose. Whichever you want to it will help you to do all the visual analogy takes That you want and we're doubling you don't have to spend much time.

Tableau will do all the work for you. So you don't have to Wrangle around the data that much you don't have to scratch your head trying to figure out how you should represent. Your data. Tableau has got lots of options available. You can choose how you want to depict it and it will do the rest for you. So this is why Tableau is popular and this is what Tableau enables you to do. So now let us understand more about Tableau by looking at the architecture of Tableau. So in the left hand side, you can see the data sources that you can connect Tableau to and to connect with Tableau.

It uses data connectors and Tableau consists of amount of optimize data connectors for databases. There are also common odbc connectors designed for any systems without a native connector and it offers two modes in support of interacting with data first, you can have a live connection or in memory connection. I'm clients can switch among alive and in memory connection, whatever they desire. So this is the analyzing part. So what happens in a live connection is that the data connectors of Tableau control your available data infrastructure, but transferring Dynamic SQL or MDX statement straight into the source database except importing every data if you have provided in a quick and analytics optimize database such as vertica, then you can get the advantages of that investing by connecting.

Live on to your data and this leaves the detailed data in the source system and sends the aggregate outcomes of query to the Tableau. And in addition. This means that Tableau can effectively utilize unlimited amounts of data. Well in fact a blue is the front and an Alex client to several of the large databases in the world and it has optimized every connector to receive the advantage of unique characteristics of every data source, and this is the visualization that you can produce using Tableau you can make it. A workbook for Tableau readers, you can also make static readers. Now you cannot work more in this static readers. This will just represent the visualization that you produced and you can also produce visualizations for web and mobile users by using Tableau server.

Now. Let me tell you how you can use in memory connection. So in memory connection is a very fast data engine to optimize your analysis. You can connect your data and after that with one click you can extract your data. To get in memory in Tableau and tableau data engine fully consumes your entire system to attain fast queries answers on millions of rows of data on commodity hardware and since the data engine can use disk storage as well. And as well as RAM and cache memory. Let me tell you that it is not confined with the quantity of memory on a system and it is not essential that an entire data set can be loaded into memory to attain its performance objectives. So what happens when a user opens a view in a client device? So when a user opens a view the user begins a session on the Tableau server and then the application server thread begins and then verifies the permissions there are security protocols defined for a particular user and then the user can have an access to the view created by Tableau.

And this is how the Tableau architecture Works in order to connect to your data source in order to analyze it and finally providing you with a visual. Data or graph or any kind of visualization that you chose to be. So let us move on and let me tell you how to install Tableau now installing Tableau is very easy. You can go to the Tableau website and download the exe file. Just run that file click on install. It takes minutes to install Tableau or sometimes it can also get installed in a few seconds. After you have installed Tableau desktop, the latest version of Tableau desktop is 10.2. So you can install that and after installing it will ask you to register to activate your version.

So you'll get a license key. You can purchase the license key. And if you are a student and you wanted for Academy purpose you can get it for free for a year. So you just have to go through the registration process and there you have Tableau ready to use. Now let us go ahead and understand Tableau little bit more. Now. Let us understand how to connect to different data sources in Tableau.

So when you open Tableau, the first thing you'll see is the connect option. Now I can connect to any files. These are the sources that I told you about. You can connect it to Excel files to text files Json files or any kind of server as well. So if you already have a data set in your system, you can just go browse on to the file. Location and you can open that. So in Tableau, you will have sample data set which is the superstore and you can rename it if you want to and when you load that data set, you'll see a preview of all the different fields and attributes of the data set that you have so you can see that these are the attributes the order ID or the date ship date ship mode customer name segments and this can all be viewed even before you open your worksheet now Tableau has also Different data types, these are the data types that will deal with there are Boolean which contains true or false.

There are dead values. There are date and time stamp values the date values just have the date the month and the year and indeed in time. There is also the timestamp in this format. This is the hours minutes seconds and AM and PM, you can also represent geographical values. So there are geospatial data when you It feels like city or state that is related to a geography Tableau will detect it and it you can represent it in geographical values.

You can create a view with maps which is very interactive. And which is also very popular in Tableau. Tableau also uses whole numbers and decimal numbers and also text and strings and all this data types are represented by symbols. You can see over here that text and string values are represented by ABC the Values are represented with the calendar icon the date and time values are represented with a calendar icon. But with the clock the numerical values are represented with a hash symbol the Boolean values artiste 5f which is true or false and geographical values are represented with a globe with latitudes and longitudes. And the best part about Tableau is that it auto detects all the data types.

You don't have to specify which data type is what but if you want to you can do that as well. You can explicitly Define if it's a number or a string. You can do that in Tableau. Now, let us see the Tableau desktop UI so when you open your worksheet, this is what it looks like so you can see there are dimensions and measures the dimensions are usually the text Data you can see it's ABC Means it's a text or it can be a date. The measures are usually numbers will know more about dimensions and measures later on this tutorial and over here with this toolbar. You can decide how you want to represent your data. You can label your data. You can use tool tips, which will help you to hover over your data and see the details.

You can include what details you want to represent by just dragging and dropping the dimensions in detail section. You can play around with colors. Colors about how you want to represent your data and this is the rows and column section. So you can just drag and drop items over here and here you can see the pictorial representation of data and this is called a canvas and this is your workspace. This is for creating new workbooks or dashboards will explore more about Tableau you I when I show you the demo so now let us move on and let us understand about dimensions and measures so Mention is a field that is an independent variable and the data types could be the strings. It could be Geographic locations numbers daytime anything and Tableau guesses the data type according to the mention names. So when you specify region it might take it as a text or it might also take it as a geographic location. You can Define that explicitly as well. So now it is representing it as text. But if you want to represent it as a geographic location, And you can specify and change the data type of these field or this Dimension and dimensions are actually used for detailing your data.

I'll tell you how but first let me tell you what our measures and in measures all the data types are numbers and these are the inbuilt data Dives the latitude and longitude. This can be used for representing Geographic locations, but mostly they're all numbers as you can see over here. And to represent a measure you always need a dimension and dimension. Like I was telling you it helps to detail your data.

Now when you see just measures let's consider sales. So you just see a number the sales is let's say 10,000 but that doesn't give you a whole picture of anything. But when you say it like the sales by region or sales by a unique product ID and this helps you to add detail in a representation of data. That is how you can get a clear picture of your data when you represented by different dimensions. And this is what is used for analyzing your data. So this is what dimensions and measures are now this is the show me data. So this pain over here shows you how you want to represent your data. So there are a number of options available. You can show your data by representing it in a pie chart by heat maps by bar diagrams.

There are different styles of representing in bar graphs. Or you can also represent it in the geographical map when you choose a data set it will automatically highlight the data that you can use to represent it. If you see over here some are blurred and some are not so if it is blurred you can understand that your data set is not compatible to use these kind of line graphs. You can use a bar graph or you can use a pie chart for that but not line chart maybe because your data that you're using is not compatible to it. So now let us move on and understand more about this visualisations. So the first ones are graph you can represent your data in bar graphs or line graphs and you can also represent both of them together. You can choose to have a horizontal bar graph or a vertical graph that you want. You can play around with colors. If you want to show different sales by different months. Let's say and you want to represent it with And graphs you can use different colors for different lines for different months.

And if you want to see two different fields, if you want to compare two different fields together, you can use a dual axis graph also so over here, you can see that we have represented this prophet and the shipping cost together. Now when I visualize it this way, this was the graph for shipping cost and this was the graph for profit. Now when you represent it together, you can see that whenever the shipping costs. Has increased the prophet also has increased so this gives you a clear picture. Right and there is a perfect correlation with shipping cost and profit. So this is how you can make analysis of your data and now you can make future predictions. If I increase my shipping cost. I will earn more profit definitely so that is why visualizations are important and you need to understand also that how you should visualize your data and the next is the geographical graph now if Viewing the sales by regions.

You can see it and a geographical graph like that. You can pinpoint the areas where you're getting the maximum amount of sales and you can also have an area graph with dual access has now this are the sales and profit dual-axis graph that you can see over here and there are many more ways of visualizing your data. Let us see some more now, this is one of the most popular visualisations in Tableau the which is called the heat map. Now colors are very important in heat map and heat Maps. The denser is the color the more value. It represents. You can see the profit if it's red save the colors get darker over here. It means that it is a negative aspect so you can see over here that this tables category. The sales is very bad or the profit is very bad because it has the darkest color of it. And when you see in case of phones, which is the lightest colors, it means the profit. And sales are very high in case of phones.

The next one is the tree map in tree maps. You can represent it in rectangular forms and also you can play around with colors over here. So the darker the green the more is the prophet and you can see in case of copiers. It is the highest and in tables it is not much and you can separate your data using rectangles when you're using a tree map now, let us understand which visualization to apply with what Kind of data sets. So in the left hand side. This is the visualization that you should choose and this represents the kind of data set that you're using. So let's say that if you're using a data set that contains this continuous values, you can use a bar graph for that and for continuous Dimensions, you should use a line graph.

And if you want to represent two measures together that we just saw and if you want it for comparison, it is preferable that you use a dual axis graph. And if you want to plot measures on geographical map if you want to see the sales by region or anything or profit by region or if there is a geographical field involved, it is better that you use a geographical graph in that case. And again, if you want to compare data between different regions or compare different to measures according to different regions, then you can use the area graph with dual accesses or basically when you have got a feel like There is a count and there is a measure or the amount of density.

Then you can use a tree map for that because with tree map you can represent the quantity as well as the density of particular measure. So you'll get that idea when you are using Tableau for a while, you'll understand which one will be better. You can also hit and try methods and you can analyze yourself which visualization will be better for your data set, but this is just to give you a brief idea because your data set. We'll definitely contain at least one or two different kinds of data among these and now let us understand functions in Tableau. Now, you can use different functions to join different data sets. If you want to combine columns, you can use joins and it uses all the SQL joins that you might have studied about.

You can use an inner join a full outer join left join right joint. And this will combine different columns together. You can also have a union to Fine Rose but the constraint or the condition is that the data fields or the attributes should be say when you are combining different rows together. And this is how you can combine different data sources together by either join or Union. You can also sort your data Accord with the Tableau sort function. Let's say that you wanted in an increasing or decreasing order you want to maybe see that which one has the maximum amount of sales and you wanted on the top so you can use The increasing order for that also and it has different sorting techniques. You can choose any sort function that you want to represent your data and set is a type of filter which we can set a condition for displaying values. Let's say that we just want to display just one kind of value. For example over here.

Let's say that if you want to represent discount, which is greater than 10% so we can use set for that and this is actually a collection of Dimension members. And you can also use Tableau UI for forecasting. So this is used for prediction. When you have a set of different Trends going on. You can represent it by a line graph to represent a trend and you can derive a future prediction or a future line graph according to the graphs that is represented for different years or different months and you can use the line graph to predict the future as well. And this highly depends on the values of graphs and the different Find some graphs that you're using to represent.

The earlier values This is highly used in the business intelligence to make predictions for Investments or different purposes. And if you want to highlight something let's say that you want to highlight a particular Trend or a want to highlight the ones that has the maximum amount of sales. So this is what you can use it. You can highlight it with a particular color blur out the other line graphs or blur out the other day.

Diagrams or visualizations that you created. So this is one way also how we can visualize your data. You can also design visuals for a particular device. If you want it in your mobile, you can select it the device type tablet or it is a mobile phone. You can do that also because the resolutions of different screens are different and you know that mobile phones are small the screen is smaller than the computer or the tablet that you're using. So all the visualizations that it created together in the dashboard. Might not fit into because it might be too small and you won't get a proper visualization so you can design it and you can tweak the visualizations according to devices as well.

So this is how you can make different visualizations in Tableau. So let us first get introduced to visual analysis. So visual analysis means gaining insights from a visual interface and in order to gain the right insight and knowledge. Knowledge from a data source we should be able to represent the data visually as accurate as possible. So what do you see in front of you is the wheel of how you can represent data visually in the most appropriate manner. So the first step is to acquire the data or to understand what data you're dealing with. Then you should filter out the data to select the correct parameters, which will be able to represent your data in a more insightful and meaningful way. Then you should take steps to enhance your data to make it visually attractive then tune it Gained the inside or to make the inference that you want to and finally deliver the data.

So this is the cycle of visual analysis or these are the steps that you should follow in order to represent your data visually as accurate as possible. And now let us understand. How do we perceive data Because unless we understand the visual perception of human beings, we won't be able to understand or we won't be able to make the right choice of how to represent our data. So now let's do an exercise. So what do you see in front of you now is Is a series of numbers I'm asking you to count the number of fives in this sequence. So let me tell you that this is the same sequence but you can see the difference. I've only bowled out the five and I've darken the color of five which was easier which took less time to count the number of fives. Yeah, so definitely the second one took lesser time because this was visually more perfect or this was visually helpful for us to count the number of files.

Whereas here you might miss out. Sighs because the letters are quite jumbled and they're like cramped altogether. So this makes it harder but when you pulled out the fives and you enhance the color of five, you can see how easy this is to grasp on the data. So this is the thing that you need to understand the how your mind actually perceives the data. Now the first one is called attentive learning. This is where you have to put on a lot of attention to find out the data or to gain insights or extract the right Knowledge from it. The next one is Very attentive. This does not require that much attention like the one before and this makes it very easier for us to grasp on. So this is the Baseline that you need to understand when you're trying to represent your data try to understand the ways and the possibilities about how you can represent your data in a more informative way.

So I hope that you have understood the purpose of giving you this exercise and to why I'm telling you about attentive and preattentive. See the more you learn with less effort as beneficial, right? So you have to find out ways in how you can represent or give more insights and extract more knowledge with just one visualization. So now here is an example. So the chart that you see so I've represent the sales of a particular company and the sales that they count are divided into first quarter second quarter third quarter fourth quarter, so they make an analysis of their sales. Dark Water so these are the yearly graphs or the yearly pie charts for the consecutive three years from 2014 15 and 16. Now I have represented this in a pie chart. If you could see these two almost looks same if I asked you that question that for the third quarter, when did this company have largest sale whether it was in 2014 15 or 16? I'm asking you for the third quarter 2014 now tell me the difference between 2015 and 2016 for the third quarter.

If you see it is very hard to detect the size. You have to observe it very minutely then only you'll find that 2015 is slightly little bigger than slice in 2016 and similarly if I asked you for 2014 15 and 16 for the first quarter. Can you tell me which one of them have the highest sales for the first quarter now this differences in the slices are very hard to detect you have to pay a lot of attention. Chin, again, this is an attentive perception. So this is another way of representing the same thing. So the gray bars over here represent 2016 sales the pink ones for 2015 and blue ones for 2014. Now if you compare and I have laid out all the quarters together to compare each of this. So like I was asking for first quarter, which one was the highest so you can clearly see that 2015 had the highest sales were in first quarter whereas in In second quarter 2016 had it in third quarter 2014 and in fourth quarter also 2014.

So this is very much easy. Now I could just look at this bar diagram and answer me in just one or two seconds, right? Whereas when you were looking at the pie chart it took you more time. So you can see that making insights from this diagram was far more helpful and far more quicker than the pie chart that we just saw. So this is the fundamental thing that you need to understand when Trying to use Tableau to represent your data because there are going to be a lot of fields that you'll be dealing with.

Your data set will have numerous amount of fields. You have to extract the right ones. You have to select the right ones with which you can represent more information in just one visualization. So now let us take a look at the Tableau product family. So the Tableau Software that you'll be using for data visualization is the tablet desktop. The Tableau desktop has got different versions that you can use. So these are the Tableau reader Tableau public Tableau server and Tableau online. So let me first tell you about Tableau desktop what you can do with Tableau desktop. So double desktop is a self-service business in Alex and data visualization that anyone can use it can translate pictures of data into an optimized database. Is it can connect directly to data from your data warehouse for Life up-to-date data analysis, it can perform queries without writing a single line of code.

It can also import data into tableaus Data engine from multiple sources and integrate them as well. You can also combine multiple views in an interactive dashboard with help of Tableau desktop. And another version of Tableau desktop is the Tableau server. Now, this is a more Enterprise based Tableau desktop. So Enterprises have to buy a tableau server. Over the large Enterprises usually by Tableau server and they use it for publishing dashboards with the Tableau desktop and they can share them throughout the organization with a web-based tableau server. What it does is that it leverages fast databases through the Life Connections. Whereas Tableau online over here. This is a hosted version of Tableau server, and it makes business intelligence faster and easier than before you can publish Tableau dashboards with Tableau desktop, and then you can share them. Their colleagues or friends and they're like we come to Tableau reader. Now. This is just a readable version. It is a free desktop application and it will just enable you to open and view the visualizations build in the Tableau desktop.

So someone else would be building it you'll just be able to open it and view it you can filter and drill down the data, but you can't edit or perform any kind of interactions. This completely depends on how the author has represented it so you cannot make any changes. But you can only use filters on that. So these are the versions that you get in the Tableau product family. And right now the latest version of Tableau desktop is the Tableau 10.2 which is the best among them. So you can directly download Tableau desktop and you can register it. You'll get a license key or even if you just want to try it first, they'll give you a trial version which is for 15 days. So you can use Tableau desktop. It has got all the features the Tableau server is for Enterprise.

Ever see if you just wanted for individual use you don't have to buy a tablet server because it is also costly so Tableau public is something where you create visualizations, but you cannot keep it with you. You have to publish it and everyone can see it. So it's up to you. I'll just suggest you to go and download Tableau desktop and I will advise you to download the latest version which is 10.2.

So we'll move on and let us see why Tableau. So these are the pros of Tableau or the features of Tableau because of which it makes Tableau very easy and very popular. So the first thing as I was already saying easy, the first pro is its ease of use will Tableau is a very interactive tool. It is very easy to learn and very easy to use you'll know that when I show you how Tableau looks like and how to make visualizations with Tableau so you can find all the options in front of you. It is very easy to represent your data. The only thing that you need to know is How to do it not how to use Tableau but how to represent your data so you have to be able to choose the right options from Tableau.

So it's a direct connect and go it means that it can connect to any kind of data source, which I already mentioned about and also you can make different kind of connections. You can extract the data from a particular data source and make changes on to it or make visualisations out of it, or you can make a live connection with it. It is also perfect for mashups it Is that you can join different data sets together.

You can include an integrated lot of data sources all together to analyze it and represent them visually because sometimes it is not enough to just take an account one kind of data set. There might be similar data sets which you want to analyze together and with Tableau you can do that and the best practices in a box. So Tableau comes in package with everything you need. It has got different filters if you want to sort your data, it has got an option for that too. And there are many other visualization. Should options that you can get with tablet. You can represent even geographical locations with Tableau. So everything that you might need to represent your data that is there in Tableau.

I'm pretty sure that there will be nothing that you say that oh my God. This is not there. It would have been useful if this was there to make this visualization look good because I'll tell you you can use shapes. You can represent it with different shapes. You can even download your custom shapes and represent it in the visualization. So that is why Tableau has become so much popular because of all these features that Tableau has so now let us learn how to use Tableau. So the first thing that you need to learn is to connect to a data source. So when you open Tableau first thing that you need to do is connect to a data source and you can connect it to any kind of file whether it's an Excel file txt file, Json file access files spatial file statistical file, if you also want to connect it to a server you can connect it.

A tableau server the Microsoft SQL Server MySQL Oracle Amazon redshift and there are even more options. Well in Tableau, you also have a sample databases. So if you're using Tableau for the first time and you don't have a data set yet, you can use their own default data set as well. So there is a super store data set and I guess there are two or three more that you can use and there are different ways to connect to your data so you can select how you want to connect with your data you can connect T' life it means that connecting directly to your data and the speed of your data source will determine the performance you can import all the data into tableaus Data engine or you can import only some of the data or the data or the fields are parameters that you might need in order to make your visualisations.

So you can choose from any of these three options. So the second two options can be also called as in memory connection and in order to make connections to a database Blew his God and amount of optimize data connectors. There are also common odbc connectors designed for systems without native connector and it offers two modes and supports of interacting with data. So as I told you the first one is life connection and the second one is in memory where you can choose to import all data or import some of the data that you might need and one great thing about connecting to data with Tableau is that you can switch among a live connection or an in-memory connection whenever you want to let me tell you what happens in a live connection. So when you try to connect a live with your data source the data connectors of Tableau control your available data infrastructure, and they transfer Dynamic SQL or MDX statements straight to the source database except importing every data and when you choose to import data, it presents a fast in-memory data engine to optimize for analytics.

You can connect your data and after that with just one click you can extract your data to get in. In oo, so what happens here is that the tableaus data engine fully consumes your entire system to attain fast queries and it can answer on millions of rows of data on commodity Hardware. So these are the ways to connect to your data and you don't have to worry about how the data connectors were. You just have to worry about how you're going to represent it. So even if I'm saying like a lot of complex sentences don't worry, all you have to do is just click on your data source, and then Tableau will Extract all the data that you want and you can start working on it.

It is just a matter of one or two clicks and when you're connecting two data sources, it doesn't mean that you can connect it to only one. I was already talking about mashups which means connecting different data sources together. And this is how you can connect different data sources together. So when you are trying to include more than one data source, it will give you an option of how you want to combine them.

So these are the joints that are available you might have seen This in the SQL joins. This is exactly the same. You can either choose a left join a right join inner join and a full outer join. So the left join means everything from the data set including the common one from the other data set the right joint means everything from the second database and the common ones with the first database the inner join means only the common part of both the two databases and full outer join means all the fields of both the databases. So this comes in very handy. Handy, when you want to analyze a bigger portion of something, let's say that at one point you are analyzing about the cricket teams of a particular country and then you decide that you have to analyze about other sports too. So there might be some feels like player name or player ID, which will be unique or this field name will be similar but the data will be unique from the two databases or more so you can use joints like this.

So the Visualizations that we have already created before and if you want to integrate the same with the different data source, you can do that using the joints. So if you go ahead and click on the show me data beIN on Tableau desktop, you can find the visualizations or the layout of the visualizations of how you can represent your data so they can be in the form of heat maps 3 maps and bargue diagrams in pie charts or even you can represent it according to geographical locations.

So you have the Liberty to choose. Is how you want to represent it? But sometimes it may happen that some of the options might be unavailable for your data and this completely depends on the data source that you're dealing with. So for some data if there are no fields for Geographic locations, you can't represent it in the geographic locations or if there is a data or the parameters that you're dealing with does not specify anything about density. You can't use heat Maps. So I'll be showing you when to use which of the maps and how you can choose the current visualization. Relation are the collection me option from the database and then there are filters now filtering means only selecting the parameters or the fields with which you want to represent or visualize your data with and there are three types of filters or you can say there are three different ways to limit the data that is going to be displayed on your graph. So the first one our SQL custom filters the second one are context filters and third one.

Our traditional filters. So what is a custom SQL filter? So a custom SQL filter is something like where there is a where clause and it is placed in the SQL that queries the data to be used in the workbook because behind the tablet dashboard on which we are working with there is an SQL query going on when you select a filter so it will specify a where clause in that so filter is a tableau term and it technically applies only to Next enter additional filters, but however, the custom SQL filter emulates the behavior of a global context filter.

So we will refer to it as such so now since I was already talking about context filter, what is the context filter? It is filter in Tableau that affects the data that is transferred to each individual worksheet. So context filters are great when you want to limit the data seen by the worksheet. So when a worksheet queries the datasource, it creates a temporary flat table and it is used to compute the the chart this temporary table includes all the values are not filtered out by either the custom SQL or the context filter. So just when the custom SQL filters your goal is to make this temporary table as small as possible. So now let me tell you about the traditional filters and tradition filter is exactly what most people think of when they think of filters. So when a tableau is creating the visualization it will check to see if a value is filtered out by traditional filter and it is not Dad the table level and it is very slow among all the filters. But let me tell you that it does have an advantage because creating a traditional filter is very easy.

You just simply have to drag a field and drop it onto the filter shelf. So I'll be showing you all in the demo where to drag your parameters or different fields from and where to drop it. So all this filtering that we are doing it is for enhancing The View that we are going to make and there are other ways for enhancing the view to the next one is representing it. With hierarchies now in order to enhance your data you can use sorting to but sorting may not always be the right option for representing your data. So while there are other methods to also enhance your data sorting can be one option but sorting cannot always be the right choice when you want to represent it. So you also have to be able to drill down to granularity of your detail to and hierarchies can provide a way to do that. So you can start with and high-level overview of data and then drill down to Levels of detail on demand. It means that you can represent even the granular form of data by hierarchies.

So if you see in the diagram over here, so there will be a category. So this is going to be the high level overview and there would be subcategories about your data to for example, maybe you are representing your data by countries and you want to go down to granular level and you can Define it by let's say states in the country and then you can drill down even more to cities and then to street names in the city. He's like that so hierarchies are very important as well. So that other way of doing it is grouping the grouping signifies the all the related fields that you can use to represent something. So for an example, let me give you the example of representing the marks card of a particular student now if it's a high school, you know that all the subjects differ in each of the standards.

So what you can do is that you can group similar subjects for eight. Then you can group the different subjects for nine standard together and thereby 10 standard also, so you can filter out the field names of the particular subjects and you can group them as class nine subjects and you can group them as standard 8 subject standard nine subjects in standard 10 subjects. So grouping also allows you to organize and manage your Fields as well and next our sets so Setzer nothing but there are a collection of Dimension members. So dimensions in Tableau are used to add level of detail onto your data and set is a collection of different dimensions together. So I'll be telling you what dimensions and measures are so you'll understand sets in an even better way when I tell you what dimensions are. So for now, you can just understand sets are nothing but a group of Dimensions which are nothing but different fields now, let us see the different data types in Tableau.

So the first one is Boolean, so as you all know what Boolean is it is either a true or false value the next one are Numbers like for example, you can see like 300 400 starting from zero to Infinity all the whole numbers. Then you can also represent decimal numbers and Tableau as well. Then you have got a data type which is called date or date and time stamp. It means that it will specify the month day year and also the time so this is represented in this format next. It can also represent text or string and this is one data type, which is very unique which is Geographic. Values so if your feet ever contains something like country City, it will auto detected the latitude and longitude measure and this is auto generated by Tableau and it will detect it as a geographic value.

And this is where you can use the geographic representation in the show me panel. So whenever you have a feel like City region countries or anything that is related to a location you can see that it will automatically generate that visualization for you. So these are the data types. You can use in Tableau. Okay. Now, let's talk about measures and dimensions.

So what is the dimension a dimension is a field that is an independent variable. It is used to add more levels of detail onto your data and it is usually a text. So whenever your data source will detect any kind of text the field that is only filled with text. It will order detectors as a dimension and if there are numbers it will auto detected as measures. So if sometimes there are numbers that you wanted to represent as a text, for example, let's say the year, you don't want to perform calculations on it. This is a year and you can treat it as a dimension or a text so you can also explicitly Define which is a dimension and measure in Tableau. So dimensions are used to add level of detail now measures are just numbers. So let's consider the same example again of a class of students and their marks.

So their Mark since they are numbers they will be treated as measures and Tensions could be the name of subjects. So if you just say a mark of a particular student that makes no sense. But let's say that you add something like the mark of a particular student in science Mark of a particular student in social studies marks of a pretty good student in literature. So that will add levels of granularity and levels of detail. So that is what dimensions can be used for. So now let us see how successful Tableau has been in the past years.

So this is Gartner's magic quadrant. Quadrant and this is the magic quadrant that Gardner created for the business intelligence and analytics platform and you can see Tableau lies among the leaders quadrant and this is the ability to execute. So this lies on top and let me tell you the Tableau has been leading in Gartner's magic quadrant since the past three years. Definitely Tableau is the winner over here. Also, let us see what different companies who have been using Tableau has got to say about it. So there are articles published. About Delight about how they're using Tableau. So there was something like the Lord builds a culture of enablement with thousands of Tableau users. The same data finality takes the finale's scores goal with data inside and they're all using Tableau.

So you can see how popular Tableau has been and still is and I'm pretty sure that Tableau will stay popular because you've already seen what can we do with Tableau. So now it's time to show you all the demo of how to use Tablo and for that we will take in account the u.s. Crime data set. So this data set contains different incidents different crime incidents. So we'll take a look at what this data set contains and we'll make analysis of this data set by making visualizations in Tableau. So let's get started.

So this is my Tableau desktop this the version is 10.2. So the first thing that I need to do, I need to connect to my database and here you can see all the options. Are all the kind of data sets that you can go connect to you can connect it to any kind of server. You can connect it to a local file that is in your system. So my file is a CSV file. So I'm going to go ahead and click on more and this is my file.

So go select update now and I have chosen a life connection. So these are the different fields in my data set. So I've got a record ID the agency gold agency name type the city the state year month incident. So these are all the different fields that we are going to work with. So it contains the crime type whether the crime is solved or not the victim sex the victim raised victim ethnicity the perpetrator, which is basically the killer that's a fancy name for a killer so This is all the killer details the weapon and it contains the crime records on u.s. Across multiple years. So we'll go ahead and we'll play around with this data set.

So next thing you need to do you need to make a worksheet where you can create all those visualizations in so here just click on this or click on here to go to your worksheet. So just click on here to go to your worksheet. So this is your first sheet. And the first thing that I want to find out from this data set is which state has the maximum number of victims or which state has the maximum number of crimes and that'll can find out by taking an account the victim gone. We can see that how many Who were killed so the place or the stage where maximum people were killed should be the highest crime state.

So this is what we're going to find out and these are dimensions all the text fields and hierarchies. So we've got the measures here which are numbers like victim count age perpetrator count and age and incident. So now I want a geographical visualization. So for that I have got my geographical measures. So these are the local values that will help me to plot the visualisations on a world map. So here I'll just drag and drop longitude on two columns and latitude on two rows and then let me select state and put it on to detail and let me just take victim count and put it on color and I can see that the colors are different some where the blue is darker and the place where the blue is the darkest is California and the victim count is 50 here next one might be Texas with seven thousand and forty eight victims.

Now, this shows the overall victim count across all years. And if you want to see it just for a particular year so we can add filters for that because we have God Dimension, which is year. So you just drag and drop here on to filter and you can select if you want it for a range of values or at least at most so you can just drag it. It and see the view across 10 years or 5 years old together. So just click on OK and then just click on show filter.

So this is going to be something like that. So I want in from 2000 to 2014. So still California has got the highest victim count with 4002 and let's say that you won't descri that is you want it to see for one particular year then just go ahead. Select any year. Let's say 1980 click on OK click on show filter. So now you've just selected 1980 now so New York has the highest number of victims in 1980 with a count of four hundred and twenty six and then comes California with 322. Then I guess Texas with 230 and Florida with 148. So this is how you can use different filters and we found out that California has the highest number of victims. So the crime rate is really high in California. So now we're going to find out that what are the weapons that are used most in order to kill people.

So we'll find out the favorite weapon of the Killer's so we'll just make another worksheet. So now I'm going to do the same thing. So I'm going to put longitude and latitude. In columns and rows, it's not again. I'm going to select victim count on two colors and then in detail, I'm going to put City. So now what I'm going to do, I'm going to change the view. Okay. So since this won't give me any kind of picture even if I select weapon over here. So I think I should go ahead and select different views in order to understand or identify that which where the objects are which were the weapons that are used the most So now I'll just go ahead and select the bubbles packed bubbles.

So this is what you see now, let us take in weapons and put it on two colors. So now you can see that all these colors which is like a teal blue. I'm not good with colors but this blue over here, which is quite different from this blue. So this blue occupies a lot of color and this is handgun and this blue represents. And again and in California it see that the victim count in California with handgun is almost 6566. So out of all the eleven thousand people were killed 6566 were killed by the handgun and apart from that. A lot of people are killed by knives also. So this blue over here this represents knife the pink one over here is unknown, which is the light pink and the Deep thing which is a Deeper color of pink this represents rifle. So the red here is Poison the purple one represents Suffocation the blackish purple one represents a shotgun.

So the green one is fire and this green one is a firearm. So even if it doesn't give you a clearer picture, you can go ahead and change the views. You can also try a tree map over here. So this is another inside that we made so now let us find out that which City has the most amount of perpetrators. Now. This is going to be very similar to the one that we find out where it has the maximum number of victims. So instead of victim count will just go ahead and select perpetrate account. So we'll do the same thing again. So again, it is showing that California has the maximum number of perpetrators, but we want to find out that whether the perpetrator count or the victim count are directly correlated or not. The one to find out that in a state where there are more victims are there more Killers or not, or is it something like there is one person who is going around and killing everybody so we'll find that out.

So in order to do that, so let's go and create another worksheet. So here let me just find out State and put it on. Columns and enroll, I want perpetrator count and I also want the victim count so you can see that there is a direct correlation, but if you want to observe it properly so you can just go ahead and do this. You can represent it in a dual axis. So now you can see that the orange ones represent victim count and the blue ones represents perpetrator count and let us go ahead and change it.

Let me just put a Line, so now I can see that the blue ones which represents the perpetrator count and the orange portions represent the victim count so you can see that there is a direct correlation. So it means that the murder scenario something like 1 to 1. So 1 Killer goes and kills one person. So there isn't much of a hint for someone to be a serial killer.

So this is one more inside that you can make from this visualization. And finally you can go ahead and create dashboards if you want to Resent it to someone. So these are the worksheets. You can also go ahead and rename this worksheets. So here it is the reading sheet. So this is the let me call it the highest victim so similarly you can go ahead and rename everything. So now if you want to include it in your dashboard, all you have to do is just drag and drop your sheets over here. So this was the one with the filters that we created. The filters will also be here then let me just drag and drop sheet number two. You can also adjust the sizes. Let me close this data pane now. There's might be some space here also, yes. Oh so you can also zoom in and zoom out it with the dashboard over here. So if you want to present it to someone you can show all your work sheets at once. So this is what you can do with Tableau. Tableau is known to create interactive visualizations that are customized for the Target demographic and what better way to learn it than a step-by-step tutorial? Hi all this is a pasta from Eddie Rekha.

And today we're going to talk about charts and Tableau, but first let me show you the topics. I'm going to cover for today. First of all, we're going to talk about the generated fields in Tableau followed by the used cases of those then we're going to talk a little bit about building charts and Tableau, which is the major focus of this session. Then we're going to talk about the pros and cons of tab. And finally, we're going to conclude our session. So without Much Ado, let's get straight into the module. Now Tableau generate some fields, which can be visible in the data pane. Now these fields are generated in addition to the fields that are present in the data set.

Now, the generated fields are measure names measure values the number of records and latitude and longitude now measure names and measure values are two Fields created in Tableau by default. Now, these fields are created when a Data set is imported into Tableau. So you can go into the data pane of the worksheet and view the fields as I'm going to show you in a little while a measure name consists of all names of a measure present in a data set and is always present at the end of the dimension. Whereas all the measure values present in a data set are kept together in the field called measure values and it is also always present at the end of the measures list it consists of all continuous values of all measures and we talked about number of God's for those of you who have worked with Excel sheets and power bi before. Your number of Records is basically like a count variable.

It shows the count of Records present in a data set. It is also an auto-generated field in Tableau, which assigns the value of one for each record present in the data set. It can be used to verify the count of Records when joining multiple tables as well apart from that we have the latitude and longitude which are basically baited with geographical detailed present in a data a data set should consist of geographical details like City country or state for this particular generated field to be used. All right. So, let's see how we can use them. Now. I'm going to be opening a new sheet in tableau. Alright, so, let's see how measure names and measure values work first. So I'm going to pick up the highlighter to show you where you can find these so here I have my highlighter.

So here at the dimensions shelf. You can find the measure names and here are the measures shelf. You can find the measure values. All right. Now in the first case we're going to be using measure names and measure values which can be used to see the aggregation of all men. Was present in a data set. Now these fields can be shown as different types of visualization in Tableau Caswell. So what we're going to do first is we're going to drag the measure names into the columns and drag the measure values into the rows. All right. Now if we turn the marks shelf into automatic Tableau automatically gives us a bar chart and if not, you can go into the marks card and select a bar chart now. This visual is created for all measures present in the data set. And as you can see, we have discount number of Records profit profit ratio quantity and sales. Same thing. We can see here. All right here we can see all the measure names and measure values moving on. Now you can do a bunch of things with this particular measure, for example, if you suppose want to delete a measure value, you have the option right here.

I don't want to delete any and also you can create an alias for measure names. It can be shown in the visualization for better. Station so we go to measure names. There's an option known as edit Alias. I'm going to select that. And in this example, I am going to give the quantity volume sales. And then click on OK and as you can see the name has changed right here in your graph. And these are just a few basic things that you can do with this if you want to analyze multiple measures in a single visual this can also be done using measure names and measure values.

All right with that. Let's go to our other generated Fields. I'm going to create a new sheet for this new worksheet and we're going to talk about the number of Records. So again for this I'm going To drag the number of records from the measure spin up 2 rows and this basically gives us the number of Records which is nine nine nine four pretty basic. There's nothing much you can do about it. But when we are going to aggregate on the bigger numbers in the bigger data set, this is something which will be very useful to us. All right, let's add another sheet and we're going to see how we can use the latitude and the longitude now as I had mentioned before these fields are associated with geographical detailed present in the data.

So you should have something like a city country or state in your data set so that you can use them now unlike other bi tools like power bi where sometimes you have to mention that a latitude or longitude is in fact geographical data here. Tableau is smart enough to Auto generate these measures. So I'm going to take the latitude here and the longitude here, okay. Let's just switch it up. Let's put the latitude here and the longitude here and you can already see a map appearing a second step will be to drag the state from the dimensions and put it on this detail present in the marks card list and this creates a geo-mapping visual as you can see on your screens right now.

You do not have to select any visualizations just by dragging and dropping your latitude and YouTube's Tableau smart enough to understand that it has to create a map. Now that was all about generated Fields with that. Let's move on to understanding how and when to build different types of visuals. Now Tableau is known to create interactive visuals for easy data interpretation. So you can create various types of graphs and Tableau based on the purpose now the different charts that can be created using Tableau and their specific purposes. Something that I'm going to show in the next segment of this session. So we're going to start with the bar chart. Let's go back to our Tableau desktop create a new sheet for this and named it bar chart pretty basic. Now. This is one of the very basic charts. All you have to do is take something on your x axis and take something on your y-axis and by default it is going to be made into a bar chart using Tableau, so I'm going to take category of product in my columns and I'm going to take see the profit into the rose.

And as you can see the automatic feature will turn it into a bar graph. And if it doesn't you can just go to the marks card and select bar graph. Alright, the next basic chart we're going to see is the line chart again. I'm going to name this sheet. It's always good to be organized because in the end we are going to use bi tool. For organization and analytics, right? So here's my line chart. So line chart basically is used to compare the data over different periods. A line chart is created by basically connecting a series of dots now these dots represent the measured value in each specific period. So again, this is pretty simple. I'm going to take the order date as I just mentioned it shows data for a fixed period of time. So I'm going to take the order date in the columns and And I'll be taking sales this time as my measure and it automatically creates a line graph.

Now Tableau is smart enough to figure out what kind of graph would you need for certain data, but even if it does not you can always go into the marks card and select the kind of graph you want next is a kind of complicated graph, which is the Pareto chart. This is basically a combination of both the charts that I just showed. So Pareto chart consists of both bar and line graph the same measure is used to create the graphs. But the measure values are manipulated differently. Now, the purpose of using a Pareto chart in Tableau is to identify the contribution of members that are present in a particular field. For example, the prophet contributed by different subcategories of a product in a retail store can be analyzed using a Pareto chart.

It can be used to show the Top members and their contribution as well. So let's try doing that. I am going to be taking the sub category of products putting it in my columns. Then I'm going to take profit and put in my Rose now stay with me because this is kind of a longer process than the other graphs, but it's pretty useful. I assure you I'm going to right click the sub category and select the sort option.

It will going to open this sort of a window. I'm going to select the descending order and then And I'm going to go to the fields and my feel name is profit aggregation some okay, then I am going to drag the prophet measure again into the rose. It's going to create two separate graphs like this. But if I right click here, you can see an option called dual axis. I'm going to select this and it's going to turn this into circles. It's basically merged the x-axis of both the measures and has Loaded it into the visualization that you can see right now.

Now next you have to go to the marks card and select some profit and as a drop-down appears going to select bar here and I'm going to go for the color a lighter blue. Okay. Now I'm going to the other prophet and I'm going to select the line graph here. And I'm going to go for the color orange. Okay, maybe a darker orange.

Alright, this looks better. Now I'm going to select the sum of profit and second one right here. And I'm going to right click here and choose add table calculation from the list. Now. It opens this primary calculation type of a window. I'm going to select running total from the calculation type because that's what we want. Right. We want a running total and then select some as the aggregation which has already selected and compute using table across. All right. Now I'm going to add a secondary calculation and this is for our second graph and here I'm going to select percent of total.

All right table across as we have done before now. I'm going to be closing this window as you can see the line graph has changed and it's not on top of the bar graph anymore. That is because we have separate primary and secondary calculations. The line graph here is showing us the total running. Some of profit as we had calculated and as you can see here and now you can basically select and change colors that you want to make the graph. Look as you like. I'm going to keep it as it is and this is the procedure to create a Pareto chart in Tableau next in our list. We have a bullet shot. I'm going to rename it. No bullet shot can be used as a gauge or an indicator to show the performance of measures now to measures can be compared to each other using the bullet shot.

For example, if you are having to estimate say actual profit versus estimated profit, we can compare both of them using the bullet shot. Now. This is going to be slightly different from the three charts that I showed before here. We're going to start with the analysis option present in the menu bar. All right, select create calculated Fields. It opens this sort of A field window and going to just name it as estimated profit. We're going to type an estimated value in this example. The profit is taken as the measures. So the calculated field is created for estimated profit. So I'm just going to type in a number. Let's just keep a 300,000. Now. The good part about Tableau is that till your expression is valid? It is not going to let you apply the changes you have made in a calculated field, which is great for beginners because then you will know exactly where you have gone wrong. For example, if I remove this you can see the field shows that the calculation contains errors.

Not just that it will even show you. The syntax of what your expression should be so here when I type a number it shows that my calculation is valid and I can apply this and there you are now go to the measures in the data Pane and you have to hold the control key on the keyboard because you have to select two different measures. So estimated profit and profit now click on this option called show me which will show you the various graphs that you can apply. Why here top right corner? This is the option is the option I'm talking about and you can see the bullet chart option also being highlighted which means we can use this particular option for the measures that we have input. So I'm going to select this and you have your bullet shot next on our list.

We have text tables. So let's just add another sheet. Going to a new worksheet to do the same thing. So a lot of this is going to be dragging dimensions and measures and dropping them into columns and rows. Don't mind me not repeating it again and again, So after we've gotten a table like this, I'm going to drag this profit into the text box present at the marks card. And here you have it. It creates a text table by default next up.

We have a heat map. Now, this is basically a graph which can visualize the data in the form of size as well as colors on different measures. Now two different measures can be visualized simultaneously using a heat map. So one measure can be assigned to size. Whereas another measure can be assigned to the color of the Heat map. So let's go ahead and create one. Now again, I'm going to be holding the Ctrl key on the keyboard and select subcategory and sales from the data pin. So let me just select these two. I'm going to go back to the show me button on the top right corner of the worksheet and select the heat map and it's going to look something like this does not have any color now. I'm going to take this profit measure and drag and drop it in the color. Now I'm going to drag let's say region. Where did the region go? You're all right. I'm going to drag the region and put it in the columns. And now this has created a heat map which can be used to visualize sales and profit across different dimensions in different regions.

Next. I'm going to show you how to make a waterfall chart which is also one of those charts which requires a little more work than the others. So, let's see how it's made my God. We've got like 13 Sheets, right? Let's rename this. All right now waterfall chart is a chart that visualizes the cumulative effect of a measure over a dimension. It basically shows the contribution of growth or decline by each member of a dimension. Now, let's take for example, you can see the contribution of profit by each subcategory by using a waterfall chart. All right, so we'll start by making a basic bar chart. So we go to a new worksheet what we had done for the bar charts thing. Take the subcategory put it on the columns take profit and put in the Rose by default. It creates a bar chart as I had mentioned earlier in this session.

Now, I'm going to right-click on the prophet present in the measure spin. I'm going to choose create and then go to the calculated field option which opens up a window like this. Now you're going to take this and do exactly what I'm doing. So I'm going to name this negative of profit. And I'm going to put a negative sign right here.

I'm going to apply this now we are going to use this newly created calculated field negative of profit into the size option present in the marks card. It's just drag and drop it. So it shall give you a graph like this after which you need to click on this some of profit present in the rows and select quick table calculation and take a running total option. Now, the reason why we are taking this negative ad hoc calculation is to fill in the gaps in our bars when we are going to turn it into a Gantt chart.

So basically we're going to turn it into a Gantt chart right now in the marks card, and now this will create a waterfall chart as you can see on your screens. Now talking about the Gantt chart. Let's see how we can create a separate Gantt chart. Now again shot is the one which shows a comparison between data in different categories. So it's basically used to identify the time taken for each process. So let's try to make one now. We're going to take the drop-down button in the marks card and select Gant bar from the list now. We're going to drag order date and put it in our columns and then Right click on it and select day now. Let's click on analysis and the menu bar and create calculated field like we had done earlier in the session. I'm sure all of you might be familiar with this window right here. You can type time for shipment and we're going to use this formula called Date difference. Now as most of you might have noticed before shows the syntax right here. We're going to put in a date part a start date and end date and start the week.

So I'm going to put the date part as day in single quotes comma in square brackets order date, which is a dimension. So it appears automatically next. I'm going to have ship date now the last start of the week you may or may not put because now that is an optional part of the syntax. So I'm going to choose to not put it and I'm going to apply it as is now I'm going to drag this time for shipment into the size part and I'm going to take ship mode and put it up in rows.

And now this is created again shot. It shows the time taken for each shipment across different ship modes. All right. Now, let's go back to something a little more old-school presenting the pie chart. Now a pie chart or something as most of you might be knowing it basically shows segment wise data. It can show the contribution of measure over different members in a dimension and the angle of the pie determines the measured value and basically it's one of the most colorful charts different colors can be assigned to the pie to represent different members in a dimension. Now we're going to do this on a fresh worksheet not going to spend too much time here. We're going to just select segment and say Tales from the data pin and then go to the show me button and select the python and there you have it pretty simple.

Let's move onto another very useful chart, which is a scatter plot. Now the relationship between two measures can be visualized using this particular plot. A scatter plot is designed by adding measures in both X and Y axes this basically shows the trend or the relationship between the measures that you select will be going to try doing that. We're going to drag discount into columns. Here's the discount put it in the columns will take sales and put it in the Rose. Now this basically creates a scatter plot by default as you can see now. I'll be taking the subcategory into the color icon and dropping it right there. Now. This basically has created a scatter plot showing the relationship between the discount and the sales for each subcategory as you can see multiple bubbles with that. Let's move on to the area chart now an error chart can represent any quantitative data over a different period of time.

It's basically like a line. Off where the area between the line and the axis is generally filled with color. Now we're going to hold the Ctrl key in the keyboard and select order date and quantity. Next we're gonna click on the show me bar right here and select the area chart icon. Now, we're going to drag the region from the dimensions pane. So that can add it in the color icon here in the area Tab and this creates an area chart as you can see pretty simple with that. Let's move onto another very basic chart called a dual axis chart. It's basically a chart which can be used to visualize two different measures. In two different chart types a date column and two measures are kind of a basic necessity to build a dual access chart the different scales in this chart help the user to understand both measures.

So again, I'm going to hold onto the control key and select order date sales and quantity. So odd date sales and quantity in the show me tab. I'm going to select the Dual combination. Ocean and this creates a dual axis chart as you can see it's pretty simple. Now you can change the color do anything you want with this chart me personally, I would like to keep it as is I think blue and orange create a very good contrast which makes your data visible and clear next and the penultimate chart I'm going to talk about is the bubble chart now above chart visualizes the measures and dimensions in the form of bubbles. It's kind of like the scatter plot, but it contributes to more effective visualization It's as simple as that. All I have to do is click on the packed bubbles option and it has created a bubble chart as you can see and finally we have a very important chart but also a very basic chart which is a histogram now a histogram shows the values present in a measure and it's frequency it basically shows the distribution of numerical data as it shows both frequency and measure value by default.

It can be used in many cases. For example, if you want to analyze The discount given by a retail shop. You can visualize the amount of discount and it's frequency using a histogram. So we're going to go to a new worksheet select discount from the measures and click on show me and this is the histogram option and this is our histogram. It was as simple as that would that I've come to the end of all the charts that I had to show. Let's move on to the features of tableau. Whoa now as discussed before Tableau is one of the most comprehensive business intelligence tools in the market right now since its Inception, it has already witnessed a steady growth and has gained a wide market share in the bi and analytics space.

So it's clearly one of the top choices in the bi space. So let's talk about a few features, which has made it gained its wide market share in bi and analytics space. So first of all, it's amazing day. Visualization Tableau bi is known to offer the most advanced data visualization options and is definitely a market leader. The users can easily perform complex data visualizations by using the drag and drop feature and it has a slick interface which is both intuitive and fast for creating customized visualization.

It is easy for any business user to create customized dashboards using the complex data and sources which makes Tableau a preferred choice for business users we have Have quality customer support and since Tableau is a fast growing company with very high customer retention ratio. Most of Tableau bi users are satisfied with the product and the technical support provided according to a survey conducted by Gartner. Tableau is ranked amongst the best bi tool with respect to customer satisfaction next up and very important. It is very easy to implement Tableau.

The rich in features is easy to deploy and upgrade as per survey conducted more than 90% of Tableau users have the latest version installed and running it clearly indicates the ease of use and upgrade next. We have data source integration now Tableau offers a simple out-of-the-box solution for integrating with the most popular data sources and analytics languages, like ironpython. They also constantly adding support for new data sources as and when the need emerges it also supports Hadoop and Google bigquery API for robust data analytics. X next let's talk a little bit about its excellent Mobile support Tableau has clearly understood the requirements of mobile users and has developed a robust mobile app, which has a very rich user interface.

It is challenging task to Showcase complex graphs and visualizations on a small Mobile screen, but Tableau has mastered this art and the visualizations adjust itself based on the screen size of the device which is being used and finally, let's talk about the rich online. This and Community now Tableau has got an active and engaging user Community which will help the fellow users to learn and master Tableau. Now the community is so huge and is so always buzzing with ideas and solutions that it has a vast vendor base who also offer installation and customization Services concluding I would like to say that Tableau is a bi tool which is changing the way we think about data it helps you harness the power of your R data and unleash the potential of it so does definitely one of the best bi platforms to choose. Hey everyone. This is ratio from Eddie Rica and welcome to the Tableau dashboard tutorial.

So by now I have given you a brief overview of whatever options will be exploring in town Loop and we have covered mostly all of the options. Now, let us take a use case of the Indian Premier League the IPL in order to understand Tableau and depth now for those of you who do not not know what Indian Premier League is. Well, it is a very popular T20 cricket match and cricket stars from different countries. Take part in the Indian Premier League. Now, there are different teams that play in the Indian Premier League. So these are the teams there is the Royal Challengers of Bangalore Kolkata Knight Riders.

Good drought Lions Delhi Daredevils the Hyderabad Sunrisers rising to New Super Joe. Kings XI Punjab and Mumbai Indians and in India. This is the most awaited and the most popular Cricket tournament ever. So we'll find out at the end of this tutorial that whose team is the best because we're going to make analysis about the IPL teams only. So let's see whose team is the best So let's understand the insights that we're going to make using a flow diagram. Now there is a huge amount of IPL data. Now, you know that every time a boulder throws a ball and the batsman that's it and scores a run. Everything is recorded. There is a record of each of the bowl that was bold each of the run that was scored by a batsman don't so you can imagine the huge amount of data that we're dealing with And not just IPL.

You must have noticed that whenever there is a match telecasted on TV. There is a pre-show or a post show that is usually held where the different experts it together and they make an analysis of maybe who's going to win. They try to predict who's going to win or who's going to lose and how much runs are they're going to score or what is going to be the outcome of a particular match now, I'm not an expert in Criminal. Ticket but definitely I can ask Tableau to help me to make all those analysis. So we're going to do the same thing. And even if you have ever watched those forth and buyer shows, you can also see that sometimes they show different kind of visualizations like the Wagon Wheel, which is very popular with shows trajectory of every bowl by a particular batsman.

So they see those visualizations and they make an analysis of it. So we're going to The same thing with Tableau only we're not Cricket experts but we sure are Tableau experts and we can do the same thing with Tableau. So we're going to feed that huge amount of data into Tableau and then we're going to make different insights. So the first thing we'll do is we'll find the orange cap and the purple cup holders throughout all the seasons. So, you know that orange cap holder is the one who scores the maximum number of of runs in a particular season and a purple cap is awarded to the person who gets the most number of wickets and that particular season.

The next inside we're going to make is who was the man of the match for a particular match and who was the man of the series across all seasons. We're also going to find the overall top five players who have performed the best throughout all the seasons. Next we'll find out the best teams across the seasons will find out which was the team that has performed the best and we'll be getting onto this insights by creating visualizations for each of them and then we'll publish this insights onto the website so you can see this is what the official IPL website looks like.

It has got the batting leaders names and what is the score card for each of the team? Now, let's have a look at the data set that will deal with so the first data set that we have is the team table. So we have a team ID and for every team ID there is a corresponding team name. So the team idea of Kolkata Knight Riders is one for Royal Challengers Bangalore. It's 2 and so on and there is also a team short code to identify each of the teams. So for Kolkata Knight Riders, it's KKR. Similarly RCB Chennai Super Kings. RCS K and the opponent ID and the opponent name are basically the same thing that we have just gone through the team ID is the same.

The opponent ID is the same field as Steam ID say it has got the same team IDs and the opponent ID has also got the same numbers and similarly. The opponent name is similar to the field team name and opponent short code is also similar to the field which is team short code now. I'll be telling you why do we need? Need the data set to be like this next up. We have the player table. This is just an example. Now. This is just a part of the data sets and there are a lot of rows in our data set. So I cannot incorporate that in this screen. So I just mentioned 24 names over here. So we've got the player ID starting from 1 and we have got player named corresponding to each of the player ID. We've got the date of birth of the particular player. If is a batsman which is Batting hand if is a bowler, which is his bowling arm and the speed of the bowler also the country where each of the players belong to and if he was an Umpire.

So if it's zero, it means that he's not next. We have got the match table. Now this contains all match details. So I've got a match ID. We've got matched date the date where the particular match was held the team ID and the opponent ID are the ideas of the teams that have played on that match. The season ID is which season of IPL was this match held on the venue the stadium where it was held which was the team that won the toss and what did they choose after winning the toss then if there was a super over if it's zero, it means that there were supernovae that means the match resulted into a tie and to break that tie they played some extra hours. If it's zero, it means that though it wasn't Hi, there was no super overplayed at if it's one. It means that was and is result this field indicates that if at the end of the game there was a particular winner.

So if it's one it means that there was a winner and that particular match if it's a Duckworth Louis now this means that whether the match was interrupted due to rain or something so that they had to reduce the number of overs and hence make a new Target. So if it's zero, it means that they'll wasn't condition like that. That and then there is the wind type whether they won't buy runs or by wickets. And how many runs did they win by or by how many wickets did it went by and this is the mouse winner ID which team ID was the winner and then the man of the match I did this is the idea of the particular player, which was chosen as the man of the match.

This is the first time buyer ID Second Empire the city where it was played and the host country and up next. We have the player match table. Now, it contains Fields like Hid player ID team ID is keeper and his captain so you can see that match ID is similar and these are the players who have played on this particular Match 3 3 5 9 8 7 and these are the player IDs of the players who played this match and there is the team ID that player number one belongs to you team number one Play number 2 belongs of team number 1 and hence 10 belongs to to an diskeeper. So this is for a particular player player. ER one if he is a wicked keeper, it will be 1 and if he's not it's going to be 0 and this is for its Captain.

It is a Boolean again. It shows that if the particular player with the player ID one is a captain then we'll be one and if it's not it will be zero and now we have the season table. So now it contains the season IDs the season a year. So the first season has the season ID 1 and it was played at It the orange cap ID was hundred. It means that the player with the player ID hundred has been given the orange cap 2 and similarly is the same with purple Gap ID and the man of the series IDs over here. So this highlights the orange cap purple cap and the man of the series for a particular season.

So these are the data sets that will deal with and I will get started with the Practical demo will perform all this will use all this data sets and we'll make the following. Insights that we talked about so here it is. This is my Tableau desktop and this is where I'll be creating all my workbooks and worksheets. So the first thing that you need to do is connect to your data source, so you have to click on connect. So the data source that I'm going to use are all CSV files so I don't see it here in the option.

So I'll just click on more. I'll browse on to the folder where I have my data set. So I want this CSV file over here, which is steam. So I'll just click on here click on open and so there it is. And these are the fields that this particular file contains. You can see that there are other data sources are there are different files over here. So this is because these are the other files that is contained on the same file location. So I need a lot of details right now. So the analysis. Is that I'm going to make is not sufficient with just three of these feelings. So I need to incorporate more data sources. So in order to do that, you don't have to do much. You just have to drag and drop the other data sources that you want to integrate this with. So the next data source that I want is the mash dot CSV here. So I'll just drag and drop it over here. Now. This will ask you for what kind of joint that you want whether you want an inner join.

That means it will have all the common fields from The teen dot CSV and mash dot CSV whether you want to left join that we have the common part of Masher CSV and team dot CSV and the entire of Team dot CSV. Similarly. Write joint is entire off Mash dot CSV with the common part of both of these and full outer join means it will contain all the fields from both this to data set. So now I have to define a joint. So I'm going to select the inner join. You also have to define the mapping it means which of The two fields that you want to match with so imagine team ID from Team dot CSV and a mapping it to team name ID, which is in match dot C is so now there it is. You can see how the integrated data set looks like. This has got all these different fields over here.

Okay. Now I want some more so I need the mash dot CSV again because I need to map some other two Fields also. So again, I'm going to choose. Inner joint and here I'm giving team ID should be equal to opponent ID. Now I need one more data set which is player match dot CSV. I have to Define team idea of Team dot CSV should be equal to T Mighty from player Master CSV. Okay. So this is done also now I'll be making an analysis during the entire season. So I'll include this to season dot CSV. So when defining our inner join, so I'm going to match the season ID from Mash dot CSV on to the season ID. season year and I need to map season ID from season dot CSV So now I'll just drag and drop this season dot CSV file.

And I'll match the season ID from match dot CSV to see the 90 offseason dot CSV. So again, I need to integrate my other CSV file, which is the player dot CSV file. and here I want the player ID of match dot CSV Here one the player ID from Clear Mash dot CSV should be equal to player ID in player dot CSV, which is there and I'm done integrating my different data sources together. So now we'll go to our worksheet and first let us make a visualization of all the orange cap holders. And as you remember that an orange cap holder is the person who scores the maximum number of runs in the entire season. Let us also rename the sheet as the orange cap sheet. Okay. So now what we need to do is we just need to drag and drop feels over two rows and columns and the fields that needs to be dragged and dropped all depends on the kind of visualization that you want.

Now for orange cap holders. I need the names of all the orange cap holders for all the seasons. I want to know who was the orange cap holder a 2009 who was a 10 11 and so on. So what do you want in our column section is the season year now, we'll find the season here in the season dot CSV. So here it is. So just drag and drop this field over here. And I'll drag and drop the orange cap ID in the row section. Okay, so here it is. So now we have got the orange cap ID and the season here and the ID needs to also tell me the name of the player. So we'll create a calculation field to find out the name of the players who actually got the orange cap. So you go to create calculated field. Let us name this calculation as orange cap calculation so it will be if orange cap ID would be equal to player ID from player dot CSV then I want the player name.

And make sure you end it and click on OK at the bottom. You'll always find your calculator feels that you have created. So what I'll do now, I'll just drag and drop this over to the Rose. In order to remove the null values over here and change the measure for orange cap calculation and I'll change it to attribute. Yeah, so the null values are removed. Now. Let us make this look a little better instead of a vertical lines over here. Let us put some little caps since we are talking about orange caps. So you can just go how you want to represent it. Go to shapes click on shape over here go to more shapes.

Go to the folder where you want to choose your shapes from now. Remember that the shape folder for Tableau is in your my documents. You'll have a my Tableau repository so you can go inside that folder and find the folder where it says shapes and there will be other folders inside there so you can copy that picture which you want to use and put it one of the folders over there. So I put it in kpi. So here is my orange cap. Okay, I'll choose this one. Click. Ah, no. Okay, and I can see there are little orange gaps over here. I'll go ahead and increase the size a little bit because they look like tiny dots yeah now they look like caps. So if you know over on each of the Gap, you'll know that 4 2008 the orange cap ID of the player was 100 and the orange cap holder was shown Marsh similarly for 2009.

It was Hayden in 2010 such intent. GE Chris Gayle Chris Gayle again, and then Michael Hussey then it's Robin the top bar and Warner and then we're not gonna lie in 2016. So here is your visualization for all the orange cap holders throughout all the seasons and similarly now, we'll create our proper cup holders. So we'll rename the sheet and we'll name it purple cap and we'll do the same thing again, wherever we have put orange now in the same Fields will put the purple cap. So you remember the first thing we did was season year and columns and then in rows you need the purple Gap ID.

Like we made a calculated field for orange cap will make a calculated field for the public app also, so we'll name it as purple Gap calculation. And the condition will also be similar. So instead of just orange cap ID will use the purple Gap ID if it's equal to a player ID from player dot CSV. Then we need to display the player name. Click on OK. So here is your calculated field so drag and drop it to the Rose section. So you can you remember that in order to remove this null values, which is the measure we're going to do the same here. And again will do the same with the shapes. So for shape, I'll choose a purple cap. Let me choose this one. Okay, so this is the increased size. So again, if you hover on you can see that 4 2008 it was so helped unveil and now piecing and so on. So now we'll go ahead and we'll find out the man of the series and the man of the match throughout all the seasons.

So at first we'll find out the amount of the CDs in each of the seasons. Let's rename the sheet as Man Of The Seas. So now what we'll do is we need to find the man of the series. So I'm going to represent this visualization in a geographical way. That means I'm going to choose the world map because every season IPL is hosted at different countries. So why not use a word map? So now you have auto-generated latitude and longitude feels so I'll just drag and drop this I'll draw longitude on two columns and latitude on two rows. So there it is. Here is my world map and I'm going to pin point out different cities and different locations where each of the matches or each of the seasons were held I put the man of the series on details over here so that if when you hover around different places you find who was the man of the series for a particular year. I'll also put filters so that you can find out that who was the man of CDs for a particular season.

So now we also have to create a calculated field to find the man of the CDs because we have the player ID, but we don't have the player name yet and a particular data set. So again, we'll call it. man of the series calculation similarly over here with say if man of the CDs ID is equal to player ID Then display the name of the player. and then end click on okay over here. So in detail, I'll add the city name so find that in match dot CSV. So this is the city name put on detail. Yeah, and let us put the host country as well. So these are the cities where the matches were played.

So there were cities in India in UAE and South Africa. So these are the three countries that have hosted IPS from 2008 to 2016. So let us know put colors. So for that I'll take them out of the Seas calculation and drag and drop it to colors. So we also change the measure. Cindy has hosted IPL a lot of times. So in the same cities there were matches played over different seasons. So we'll add some filters so that we can visualize it year after year. So in filters first, let me put the man of the series id Sometimes when you can find different fields just use this search over here. So I want the man of the series idea.

Okay, so here it is drag and drop it to filters. So select all of them click on okay. I'll select this class select all except null because we don't want null values. And let us also put the season year as a filter because we need to search who was the man of the series for a particular season. So this is the season year and let me drag and drop it to here. Just click on show filter. So now you can see that this is for a particular year where if you just hover on the top of a circle, you can see that the man of the series was Watson. So you just have to drag this filters over here if you want to see let's say in 2009. So in 2009, it is Adam Gilchrist. Similarly. You can go on check for each of the individual years.

So let's check for 2010. So again for 2010 it is such in Tendulkar as the man of the series. So here you can just drag over the years to see who are the Mount of the matches you could also choose to represent it in a more simpler way not in a world map if you want to so for man of the match, let me just show you a very simple visualization because sometimes keeping it simple is the best So we'll call it man of the match. So here we'll do a very simple visualization one just use columns. So we'll just use a rose over here. So what we need is we need the match ID. And then we want the man of the match. Okay. So let us also create a man of the match calculated field because we have to display the name of the player.

Okay. So here I'm getting updates about the IPL. So it's a good thing that I am making analysis of the IPL details from 2008 to 2016 and the time when there is IPL going on in 2017. So if I could have waited a few more days to get it over I could have included this year to so K. So let me just close this. Let's get back to tableau. So we are supposed to call it man of the match calc, right? So again, it's a similar thing. So we'll say that if okay, let's not searched by this man of the match ID. Will be equal to the player ID then display the player named just like how we did it. So click on ok now we'll drag and drop this field, too. So let us also include the city name. Get there it is. So in the text part, let me include something. Let me include the wind type over here so that this column would get filled up.

Okay, so the team which was a winner they won't buy runs or by Wicked so you can see it displayed over here. So let me remove the null values. Okay, so we'll add some filters also over here. So just drag and drop this one at filters to so we'll remove the null values from here you can okay. So now you'll find that particular match ID who was the man of the match? Where was it played and the winning team did they win by wickets or by runs? So this is a very simple way to represent your visualization yet. It gives you a lot of information as well. So you can add more filters here if you want to but you can leave it like this if you just want so this was the workbook for the orange cap purple cap man of the series than man of the match.

So now what we will do is we want to see the overall best team that has performed the best entire all the seasons. So for that will create another workbook because our data source would be a little different. So let me just save this one So I just name it IPL WW1 and now we'll create a different workbook for that just file click on new. So here I need to connect to data first. So click on more I wanted the teen dot CSV again. So here it is Dean dot CSV and I'll integrate this with the mash dot CSV.

I'll Define an inner join where my team ID will be equal to the opponent team ID. And then I need the team dot CSV again. So drag and drop it. So integrate this with the mash dot CSV, so I need team ID. From match dot CSV. Just let me search from Mash dot CSV and I'll select the team from Team CSV one. Okay that there is no again. I'll choose team dot CSV and drag and drop it here or choose an inner joint. I will again choose match dot CSV, so I need the match winner ID. Should be equal to the team ID. Okay, so here is my data source, so we'll go to the worksheet. So we'll rename it as the best team. So now what we'll do we'll just again drag and drop some Fields over here. So we need the match-winner ID from Mash dot CSV. That's here. We need the team name from Team CSV to Let's get the team ID from match dot CSV.

There it is. Then we need the city name. And we need the opponent ID search for the opponent team it there it is. So I'll just drag and drop this over here. So in colors, let me put team name. from Team CSV to so over here in the column section first thing I'll do is So in order to represent it in a better way, I'll click here. in measure cell go to count distinct. Then I'll add a quick table calculation. Let's make it running total. Let us modify this more. So we'll right click this one and we'll use specific dimensions. and restart every match when i d so we'll close this and let's make sure it fits the height. So after you select the fit height, you can see the performance of all the teams together and you can see that Chennai Super Kings has got the highest peak over here. So the second might be the RCB the Royal Challengers of Bangalore with 41 and the score for cska is 44 and here is the Reston Royals with a score of 40 and close to Rajasthan Royal is the Delhi Daredevils at 39.

So You can see all the teams over here Gujarat Lions Rising Pune Sunrisers Hyderabad Kochi tuskers Pune Warriors. You can see that the highest peak is for Chennai Super Kings. So they have performed the best. They have one most matches throughout all the seasons. So this is how we have a visualization of the best team. So now we'll go ahead and we'll find the best players throughout all seasons. So we'll do that in a different workbook as well because the data source is again going to be different.

So we'll save this one. so we'll call it call it IPL W-2 I'll go to file click on new. So this is my new workbook. The first thing I need to do connect to data go to more now. I have a data set known as Ball by ball. So I'll select this data set over here. And I'll integrate more data sets with it. So the first one will be matched or CSV. Now here, I'm matching match ID of ball by ball to match ID of measure CSV, it's fine. Then I need player dot CSV drag and drop it. I'll put the striker idea over here and I'll map it to the player ID. Okay, that's done now. I need the player match dot CSV. No, I hear I want the match ID should be equal to the match ID in the player MedStar CSV. Now it automatically got mapped now next. I want team dot CSV drag and drop this and hear what I need to do is I need to select a field from Ball by ball CSV and I'll take the team batting ID.

And here I'll map it to team ID from dot CSV. Okay, so here is my data source, so we'll go to our worksheet and we'll rename this sheet and we'll say top players. Okay. So again, I'm going to go ahead and drag and drop the different fields. So the first thing I need is the player name. So search player name over here. Here it is from Pluto CSV. So here is my player name now. I want this Striker ID. from bulb eyeball CSV I'll select the team ID then I need the team name. the season ID and my chai tea Okay, so we'll add some more filters. So we'll drag and drop this player name. Over here we'll select all will click on okay. We'll also put this in colors. So now what we'll do we'll select the batsman scored into the column section. Okay, so we'll do a running total lat quick table calculations and running total. So now what we want is we want the top batsman's the top five bathrooms.

Let's say so we'll add a season filter to it that for a particular season who were the top batsman. So so we'll create a calculated field. So Let me call it. top five calc and this is it. Let us first choose season 8. That means last year. Let us choose season 8, this is for 2015 season. So we'll click on OK over here. So we'll go there right-click will go to edit table calculations. Let's do the same thing over here. All right. So we'll also add the season ID filter over here. So let us find season ID. Okay, since we added the calculation for the season 8 so will unselect everything. And let me sort this one. So now we'll just list out the top five players. So we'll go here at it filter great on top by field. Let's select top five.

And we have made that calculation which is known as top five calc, right? Here it is. And click on okay. Okay. So now if you click on fit height you can see that these are the top players in 2015 a be de Villiers. Rohani Warner Simmons and Koli so you can hover over here and if you move your cursor, you can see this track the running sum of the batsman was 562. For coolly, it's 505. That's 544 Simmons. This also 5:14 for Warner and it's 532 for de Villiers. So these five were the top players according to our visualizations. We made we use filters to find out this top five players in 2015. So let's go to the actual website and verify if these are all correct. So go to my browser over here. This is the official website. Okay, so days are the 2015 batting leaders? So we'll view the full list. And you can see that we have found our top one.

So we had Warner with 562. If you remember that then we had Simmons and EB de Villiers and without Coley. So these were the five and we got that five. So yeah, there it is. So it means that we have made all correct insights. Now, what you can do is that you can copy all the sheets worksheets that you have made and you can include it in a dashboard. So now we have created all the visualisations that we need in order to make Insight but they are all in different worksheets.

If you want to view them all together or maybe you wanted for presentation purpose you want to show it to your senior your manager who can view the entire thing at once so for that you need a dashboard. In order to create a dashboard you just go and click this icon over here. And this is your blank dashboard.

So these are all the worksheets that we have created. So what we can do we can just go ahead drag and drop all these worksheets over here so that we can have an entire view of all the analysis and all the visualization that we made using Tableau. So I'm just going to drag and drop these sheets over here. Now I'll just drop the purple Gap. You can also adjust the size. There you go. And you can go ahead and add all the different sheets as well. So this is how it is going to look like so now I've adjusted all the sites and I've made the dashboard already. So this is how it looks like so I've got all my worksheets here, whatever. I want to view. It's right here.

So this is how you can create visualizations and put it all in a single dashboard. So that everyone can see all the worksheet and all the visualizations that are created at once. So this is what you can do with Tableau. This is how you can create dashboards. You can add more sheets to it and you can adjust it in different ways also. Using functions in Tableau is essential for being able to represent your data in the best possible way Tableau luckily has a list of functions that you can directly apply to your uploaded data. Now if you've used all the functions such as SQL or Excel these functions should already seem very familiar to you. Hi all this is a pasta from at Eureka and in this module. We are going to talk all about Tableau functions now in Tableau a user can We use different types of built-in functions, which can be applied to the following kinds of parameters. So before we begin, let's look at the different categories in which we are going to be using these Tableau functions.

So we have your number functions followed by string functions. Then we have date functions type conversion aggregate functions. And finally we have logical functions. So without Much Ado, let's get straight into the module. So first of all, we have number functions, Number functions allow you to perform computations on your data values in the fields. So these functions can only be used with the fields that contain numerical values pretty obvious right on your screens right now. You can see all the number functions that I'm going to be covering in this segment. We will start with the absolute function or the ABS and close using the Z and functions Each of which I'm going to demonstrate to you using the Tableau desktop.

So let's get started. But with the first function, so the first function in the number functions category is EBS or the absolute function on your left. Most column. You can see the name of the function in the column in between you can see its syntax and on the right you can see the description. So what your abs does is it Returns the absolute value of the number or the parameter that you put in the bracket? So if I give like a negative number in there like a minus 5 then an absolute function is going to turn. To into a 5 and return it back to you. It's kind of like the mod of a number in the example that I have given you can see we are trying to get the absolute value of budget variance. So suppose you had a field called budget variance. The ABS function will return the absolute value for all the numbers that are contained in the field.

So let me go to my Tableau desktop and I'm going to demonstrate how this works. So this is my Tableau desktop and currently I'm using the 2019 point one version. This is the show me bar. If you want to know more about this interface and how to use Tableau desktop. We already have a few videos a few tutorials in our YouTube playlist. Please feel free to go ahead and check that out. This tutorial is specifically for functions in Tableau.

So we are going to die straight into that. So as you can see I have already pulled in the data from the sample data set that is available in Tableau. It's called sample Superstore. Now. This is only available in the desktop version and not in the public version. So make sure if you're trying to follow through with this tutorial you are using the actual Tableau desktop. So what I'm going to do is I'm going to navigate to analysis and create a calculated field this what you see is the Ation, editor that opens you're going to name it absolute as we are going to try the absolute function right now and I'll show you something simple in the beginning.

It's as simple as this you're going to go ahead and do abs and suppose I put in minus 5, which is the number at the bottom. It shows my calculation is valid, which is a great thing about the calculation editor in Tableau. It prevents you from making further mistakes helps you correct them immediately because the bottom it Going to prompt you whether your calculation is valid or not. Then I'm going to apply it and okay. So basically if I bring this here, let me do the attribute.

Yes, so it will give me the absolute number instead of negative of 5, it's going to give me five because absolute is kind of like your modulus. It's going to give you the absolute value of the number that you have produced. So let's try it with a field that we have here. So Let's go down. So we have this sales field here. So let me remove this and you can go make some changes. So I'm going to edit this field and instead of placing just the number. I'm going to put in sales. It's automatically going to appear to you. That is how smart this tool is. All right. So when I bring this here, it's going to automatically show number here. Now. What I'm going to do is I'm going to divide this by state.

All right, so we can see all the absolute values over here. Now, what I'm going to do is I'm going to go to show me and put it into a map. So now we can easily go to each state and see the absolute sales in all of these states. We have South Dakota North Dakota Montana and all the other states as you can see it shows us the absolute value of the sales in that particular State. All right, and as you can see on your right, it will show you the gradient of color which basically means that the lighter the color is the lighter blue. There is the less of the sales are in that particular State as opposed to the darker the color as you can see in the state of California. Or it may be New York. The signals are the maximum no rocket science there. Let's go back to our presentation and look at our next function. So our next function is the arc cosine or ecos it basically Returns the arc cosine of the given number and the result will always be in radians.

So if you give an arc cosine of minus 1 you supposed to be getting something in lines of one four one five nine two six as simple as that. So let's go back to our Tableau desktop and run this and see again. We're going to go to analysis and create a calculated field. Your Google and type equals minus one.

And apply it and like the absolute measure that we had discussed about the previous time. This also appears under measures in the data pain just like your other fields and you can use it in one or more visualizations. So I'm going to bring this here. And it shows three point one four two, which is just the rounded-off version of 3.14159. And that was all about Arc cosine. Next we have arc sine, which as the name explains Returns the arcsine of the given number and Radiance. I'm quickly going to show that. here again And it gives you the answer one point five. Seven one. Next we have the arc tangent, which is given in radians again. It is a tan which is the number function. I will go through this section pretty quickly as these are not the functions which are used very regularly in Tableau, but they were kind enough to provide us with these functions.

So we are going to use them. So I'm going to make an edit in this again. Just quickly going to go through. And you're supposed to be getting 1.56 5. Next we have ceiling now. Basically what it does is when you pass a decimal number to ceiling it is going to round it off to the nearest integer of equal or greater value. So if I give it three point two, six or something, it's going to round it off to straight up for now. These are available in a couple of data sources like Microsoft Excel and text file and statistical files, but they are also not supported by a few.

Popular sources such as Microsoft Access and action vector or Amazon Aurora so on and so forth. So here again, I'm going to go to create calculated field and I'm going to name it ceiling. Ceiling and then I'm going to put on 9.12 6-5 just any number. So as you can see the ceiling what it did was it rounded off that number to 810 right? Let me try to So that's all about ceiling next we have caused now cause or cosine of an angle. It's pretty self-explanatory. Most of you I'm sure would have studied this in school it basically Returns the cosine of an angle if you give your angle in radians, so what I'm going to do again, so as you all might have studied in school cos Pi by 4 is 1 by root 2 or point 7 0 7 so we can apply this.

Alright, so as you can see it gives you point 7 0 7 1 which is 1 by root 2 which is again caused by by for next you have caught or cotangent of an angle same as cause that you're going to I repeat next you have cotangent or cot. And we are going to implement it the same way as we did cause I don't think these trigonometric functions need a lot of explanation. We've all done these in school. So Just get this over with. So again, I'm going to go with the pi by 4 because I'm lazy and I'm not going to think of any new angles. So as we all know caught by by 4 is 1 and as you can see the value immediately changed to 1 the next function we have is degrees. Now what this does is it converts a given number in radians to degrees. And as you can see Pi by 4 as we all know Pi is 180.

So your Pi by 4 naturally is 45 degrees and that's what it was supposed to do and that's what it did moving on. We have div or division. So what it does is it's going to return you the quotient where your integer one is, basically your dividend and your integer to is your divisor and it's going to give you the integer part of your shouldn't so if I do 11 by 2, it's not going to give me a five point five, but only if I've as you can see your quotient is 5 next you have the exponent or exp. Basically. It returns erased to the power. You're given number. So basically if we start with something simple, like we'll just put on a digits a arrays to it straight up going to give you erased to do which is seven point three, but you can also use these to put in fields and formulas such as your growth rate into time.

So on and so forth next you have floor now, this is kind of the counterpart of the ceiling function where it rounds the number to the Nearest integer of the Lesser value so like and ceiling if I had to put in 3.1415. It would give me a for a floor on the other hand would give me a 3 so as you can see I have used ceiling right here. So, let's see. What have we done for ceiling. Okay, so I had put in the number nine point one two, six five four ceiling now. Let's try putting in the same number for floor. So I'm just going to make an edit in this Going to put floors. Okay, apply and OK and as you can see immediately, it went from ten to nine. Our decimal has been rounded off to the lower number. Now again, like sealing it is not supported by Microsoft Access your action vector and your Amazon Aurora redshift so on and so forth, but it is supported by your Microsoft Excel and text file and statistical files.

There are many other files which support and also do not support ceiling and floor, but popular ones like Google analytics and Google big query support. Both of these functions. Next. We have hex bin now, basically what it does. Is it Maps an x and y coordinate to the x coordinate of the nearest hexagonal bin. Now these bins have side length 1 so the inputs may need to be scaled appropriately now the heck Spandex and the hex Why are binning and plotting functions for hexagonal bins now these bins are efficient and elegant options for visualization of data in the XY plane such as your Maps or visualizations dealing with geographical data on your Scatter Plots your hex plots now because these bins are hexagonal each bin is closely approximated to a circle and hence, it minimizes variation in the distance from the data.

Point to the center of the bin this makes clustering both accurate and more informaiton of next. We have the natural logarithm of a number basically you type Ln and then your number and it's going to return the natural logarithm if your number is less than or equal to 0 it's going to return a null value. As we know that the natural log of 1 is 0 so that is what we get. That's the only one I remember from school. And that's the reason I put that in as my input. Now something very similar is logarithm with a base which we usually take as 10. So if the base value is omitted you decide to not put on a base. It's by default going to use base 10. So we're going to put in say a hundred and ideally we supposed to get to and that is it log hundred base 10 is 2 because trendiest to do is hundred pretty simple.

Next. We have Max which as the name suggests. It Returns the maximum of the two arguments which are passed. Now the arguments must be of the same type and this function returns. Null if either of the argument is null so mad. can also be applied to a single field if it is used as an aggregate calculation, which I'm going to cover in the later segments of this particular session, so It can be as simple as say two numbers. So if I put it right here show by attribute 7 is larger number than 4 so here you have seven or you could go ahead and put into Fields over there. Let's say profit and sales. and when we Take the sum of it. Now this I remember is the value of sales obviously sales is going to be more than the profit. The sales field is going to have a higher number than the profit. So that's about it. Now the other half of this particular set is the minimum function which is like the maximum function, but it Returns the minimum of the two arguments of the same type.

So if I just went ahead and edited this particular, Function instead of Max let's name it men and here again instead of Max. Let's name it men. Let's put in the Min function going to apply it and this I'm sure is the prophet field as you can see. The number has changed next we have pi as the name might suggest it is going to return the value of pi. Three point one four two, which is the rounded off value of pi next we have power. So again, we are back at our Tableau desktop, and we're going to try and Create a power field. So here I'm going to just go for power and then our base which let's take 10 and then the power so we'll take that as 3 and ideally our answer should be a thousand.

So if I drag it to Rose Bring it to its attribute. We can see pauses equal. Mm. Next we have radians which basically does the opposite of degrees it converts the number given in degrees to radians. So I'm going to go here again. So earlier we had converted Pi by 4 into 45 degrees. So here I'm going to put 45 degrees and let's hope we are getting Pi by 4. So I'll take this rad calculation put it here. end we get 0.785 for which was also Pi by 4. You can do the math use the calculator take this moment pause this you can go check then we have round. So basically what this does is it rounds a particular number that you put into a specified number of digits, right? The decimals argument specifies how many decimal points of precision to include in the final result now if that part you decide to not fill anything in the number is rounded to the Nearest integer right now before I move on to my Tableau server some databases such as SQL Server allows specification of a negative length where negative one rounds to the tens place negative two rounds to the hundreds place and so on.

This is not true for all the databases, of course such as Excel and access do not follow through with it. So with that let's go to our Tableau desktop and see how this functions So what I'm doing here is I'm going to round up every sales value to an integer. That's why I did not put anything in the decimals place. If I take a subcategory, all right, as you can see everything has been rounded up to an integer no matter what the sales in dollars is there might be some decimals in there. But all of it has been turned into integers using the round next we have sign which Returns the sign of a number now the possible return values are negative of 1 if the number is negative and zero if the number of 0 and 1 One is if the number is positive. So I'm going to run this particular example, which I have given here.

I'm going to run that itself on my Tableau desktop. So I'm going to be taking the average of the Prophet field for fat and apply. Now because the profit is positive in our given sample set. Okay, this looks better. Now. What was I saying? Yes now because our profit is positive in the given sample set our answer turns up to be one if it were negative. It would have been a negative of 1. All right, after sign we have sine which is the trigonometric sign again. So it basically returns to you the sine of an angle or we have to do is specify the angle in radians.

So let's just So again, I'm going to take sine Pi by 4, which is 1 by root 2. So it comes up to be 0.707 same as your cosine, then we have the square root of a number which as the name suggests Returns the square root of a number. So as we can see the square root of 100 is 10 next we have the opposite of square root, which is the square of a number pretty self-explanatory. I'm just going to edit this. Name of square. I'm going to do the simplest thing and you might just think this is a lazy woman, which you're right, my friend, so I'm going to square 10 so we can get a hundred and there you have it. It's a hundred no big surprise there. Would that we have tan. Okay. So here we have the sign what I'm going to do.

I'm just going to edit this not going to create any extra sheet here. Just going to put Dan and change this into 10. Tan Phi by 4 is 1 for those of you who do not remember from school. And here we have the answer as 1 and with that it brings me to the final of the number functions. We have ZN which Returns the expression if it is not null. Otherwise, it returns zero now, we basically use this function to use zero values instead of null values. So this is how it works. I'm going to put in a field. Which I know does not have a zero value. So I'm going to put in profit.

So here I'm going to put in the average of profit. It's in the some more now. I'm just going to take the average and I know for a fact it's not a nonzero value and its return to me the average of profit which is twenty eight point six six with that. We come to the end of the segment that we head back to the presentation to start with the next module. The second segment is string functions, basically string functions allow you to manipulate string data. Now you can do all sorts of things with it like pull out all the last names from your customers. To a new field one member might be say shubham Sinha and then you can pull out all the customers with the surnames Sinha and then put them in a new field using the string functions. So again on your screens are all the string functions that I'm going to discuss which is all the string functions that are available in Tableau this by no way is an advanced tutorial. I'm just going to show you how each syntax works.

So let's start with the first one. And we have a ski as most of you know, there is an ASCII code for every character of the string and what this function does is it returns that ASCII character? Now as we all know the stringy has the ASCII code of 65 and that is what this returns next we have care which is kind of like the counterpart of ASCII here. It's going to return the character that is encoded by the ASCII code here. You have to put in the number and it will return to you the character. and as you can see, it returns a next we have contains which basically returns true if a given string contains the specified substring.

And as you see it returns true next we have ends with which returns true if the given string ends with ascertain substring. next you have find which Returns the index position of the certain substring that you're trying to search for in the string or zero if the substring is in found if the optional argument which is start as you can see at the end of this argument bracket is added then the function ignores any instances of the substring that appear before that index position start the first character in the string is position 1 As you can see it's said that the substring EK that I was searching for starts from index number 5 now, let's see what happens if we put in our last argument.

Let's go to 7. As you can see because till the 6th index it was not asked to search. It returns a 0 which means it could not find the substring next we have find in it which basically Returns the position of the anyth occurrence of a substring within a specified string. So n is where the argument has occurred. Secure and find an it. I've changed the word. I've put in learning here and I'm going to search for when does n the alphabet N occur for the second time in the word learning.

All right, so if I bring it to text you see the answer 7 because L EA R ni NG. It's an 8-letter word where the second time and occurs is at the 7th position now left Returns the leftmost number of characters in the string. Ring so you put in the string and then you put in the number till which you want your substring to be. Let's see. What do I like? I'm going to put it in Matt dough and then three. So it's going to print the left-most three characters from the string. Then we have a pretty common one called length.

If either of you have used Excel or word or access or any of these common databases and apps you know, what alien does it returns to you the length of the string? So what I'm going to do? Is there you go. It Returns the length of Matador which is 7 then you have a function called Lower basically you put on a string here and it will return you the string with all characters in lowercase. So I'm just going to randomly capitalized. Okay. So next we have Ultram now this basically Returns the string without any spaces on the left. So if I try to L trim space matadors space it is going to remove the space at the left hand side of your string and here you can see the space on the right is there with the space on the left is gone.

Then we have Max which Returns the maximum of A and B, which must be the both strings. Now this function is usually used to compare. Numbers but it also works with strings. Basically what it does is it finds the value that is highest in the sort sequence defined by the database for the column. It returns null if I the argument is null. So I'm going to put apples and bananas just expression 1/2. You can put anything. And then it returns bananas because obviously more number of characters next we have mid which Returns the string starting at the index position start. So the first character in the string is positioned one if the optional argument length is added then the return string includes only that number of characters. So what I'm going to do is I'm going to use the word learning. And then I'm going to start with the second index number which is supposed to be e and then I'm going to give a length say five. So five characters, then I'm going to apply it and okay and as you can see five characters starting from position 2 next we have Min which is the counterpart of Max it basically reverses what we got and Max it's going to give you the string with the minimum number of characters.

So if you're going to do the same thing apples and bananas. You can also put in a field in here and it'll give you the least value from that certain field. So as you go apples having the minimum number of characters, then we have right which Returns the right most number of characters that you have specified in your syntax. So if I put in learning and for it should ideally give me an Ing and there we have it. Then we have the art room which is kind of like the L trim but from the right, so if I put in a string with spaces on the right, it's basically going to clear out the spaces. You're going to type the same word spaces on the left Android both and when we get it you can see there is a space on the left but not on the right, which is the opposite of L trim where we deleted the space in the left and we kept the space and the right next you have space which basically returns a string that is composed of the specified number of repeated spaces.

So space one is going to give you one space to was going to give you two spaces so on. So forth then we have split now. This is an interesting one. It Returns the substring from a string using a delimiter character to divide the string into a sequence of tokens. Now, the string is interpreted as an alternating sequence of delimiters and tokens. So, let's see how this works. and I am going to C and T Let's put these quotes on both sides. And then I'm going to put in what my delimiter is which is the – now I could have also used. / the only thing you have to do is specify it in the delimiter part of the syntax and then I'm going to put in the token number two, which otherwise should have been my delimiter this one the – but because we specify the delimiter supposed to be returning a b and as you can see it returned be now the split and the custom split commands which are Kind of the same commands are available for if you data types and they're not available for a few data types.

So oo data extracts Microsoft Excel text file PDF file Salesforce all of these support the split and some data sources impose limits on splitting string and there are these ones with support- tokens where there is a limit on the number of splits that is allowed per data source next we have trim now. Basically trim is the combination of L trim and or trim it is going to remove spaces from both left and right. I'm going to again use as you can see space and matador and another space.

And as we can see, it has no spaces in the beginning or the end and here we are at the final string function, which is upper which Returns the string with all characters in uppercase. So I'm going to go here. And go with upper. And randomly capitalized like I had done earlier. Let's see. And it returns to us Matador in all caps. So with that we come to the end of string functions. The next segment is on date functions. Now as the name suggests the date functions allow you to manipulate dates in your data source such as your month date date or time. So let's go to our first function. We have the date add now the data are returns a specified date with the specified number interval. Added to a specified date part of that date. For example, if I put in month as the date part 3 as the interval and it date as the date this expression would add three months to the date 15th of April 2004.

So let's try this. You know this expression adds three months to the date. So we had zero for as an April for this date. But if I change this to month it has July, which is May June July edition of three different months. Next we have date diff which Returns the difference between the two dates that you have input into this function. The start of the week parameter is the one which you use to specify which day is to be considered the first day of the week.

It is optional, of course, but in a lot of countries you consider Monday as the first day of the week, whereas in the other part of the world you consider Sunday as the first day of the week if this part is omitted then the start of the week. Is determined by the data source, so, let's see how that works. So as the date part, I am putting in we then I'm going to put in 2015 see 2015, October and let's see 23rd.

And then I am going to put 2015 and the same thing on the 11th month November. I'm going to put in the start date as Sunday. Alright, I think with that we're done. And we're going to change this. And it shows that the difference is 5 weeks. Next we have date name which Returns the date part of the date as a string and the start of the week parameter again, like the previous function you can use to specify which day is considered the first day of the week. It could be Sunday or Monday or Tuesday depending on what you like. So if I put in the date part as ear and I put in the year and the starting date of the week, which in this case is not really necessary. It returns a string 2004, which is the name of the earth that we had put in next. We have date part which kind of does the same thing as the date name, but it Returns the date part of the date as an integer instead of a string.

So if I made that edit here. It was returned to me the same thing but it is not a string but an integer next we have date trunk. Now as a lot of you might have guessed this basically truncates the specified date to the accuracy specified by the date part. So basically this function returns a new date, for example, when you truncate a date that is in the middle of the month at the month level this function will return the first day of the month again.

This has the star The week parameter which you can use to specify which day is considered to be the first day of the week. So on and so forth. So I've put on this date and I've specified quarter. So it's supposed to truncate this part of the date completely and bring me to the beginning of the month. So when I bring this here So as you can see, it has brought me to the beginning of the month July and has truncated that part of the month from half the month 15th of July to the first part then we have day which Returns the day of the given date as an integer. As you can see it's 15 next we have is date. Now this returns true if a given string is a valid date. so if I put in February 30th 2015 it returns false because there is no 30th in February on the other hand.

If I go ahead edited and put in like a 28 to Feb 18 turn stru because 28th of February this tin every single year. Next we have make date which returns it is value constructed from the specified your month and date. So if I make date with 2015 3 and 15 I'm supposed to be getting 15th of March 2015. This is a slider motive of the previous function. This is basically an extension of make date time and along with the date. It adds the time as well. Next we have make time which is exactly like make date except for its for time. So it returns an arm minute and second off the time that you put in now going to spend much time here as they are all pretty similar in nature. Next. We have Max that we have. Two times before so I'm not going to demo this is well again, you will put into dates and it is going to return the maximum of the two dates.

Next we have men again, you're going to put in today. It's and it's going to return to you the minimum of those today's next we have month which Returns the month of the given date as an integer. So if I put in 2005 0 9:23 I'm close this. It gives me back nine which was the month. I had put in. So this is an interesting one. This is called now this Returns the current date and time. Now what it returns varies depending on the nature of the connection. So for a live unpublished connection now Returns the data source server time and for alive published connection now Returns the data source server time, so I'm assuming that it is going to give you my system time.

So just going to go with now. I gives you 2019 if I bring it down to the date, it's going to show 7th of May 2019, which is the day in which I'm recording this video. Another one like now is today and basically what it does is it returns to you the current date? So instead of now we're going to put in today and it gives you today nice isn't it? And finally the last date function we are going to do is you're so basically it Returns the year of the given date as an integer.

And if I bring this here, it gives me 2014 which is the year. I had put in and it's given it back to me as an integer. Next. We have type conversion functions. Now, why do you use these functions now that conversion functions allow you to convert fields from one data type to another. So basically you can convert numbers to Strings such as age values so that your tablet will not try to aggregate them. So let's move on to our first type Version function the first one is date now. Basically this returns a date given a number string or date expression. So if I go in And put an order date. Basically has given me a list of all the order dates that are there in the data set. So I can obviously filter this by ear. And bring it down to these many dates. Next we have daytime which is kind of like date but it returns a date. I'm given a number or string now the some number show you the demo because it is very similar to the date function.

Now we have date parse. Now what this does is it converts a string to a daytime in the specified format. Let us that appear in the data and do not need to be par should be surrounded by single quotes and for formats that do not have delimiters in between them suppose. Fins and slashes and dots verify that they are passed as expected. So the format must be a constant string not a field value and this function returns null if the data does not match the format.

So, let's see how this works. So this is my format. And this is my date 15. September and 2005 And this is exactly how it's going to show it to me. This is the same in the table next we have float. So basically if I enter a few 3 it is supposed to give me 3 .00 next we have int or integer which again opposite of float going to cut down the decimal part of the number and give me only the integer part. And finally we have a string expression. Basically you're going to pass in an argument through it and it's going to convert it into a string and give it back to you. So that's all about type conversion.

Let's go back here for the string. I'm going to give you a demo. Rest all is kind of the same how we had done before and how you do it with other tools. So basically what I'm going to do is I'm going to edit this and SDR and I'm going to put in the postal code. Now as we all know postal code is in numbers. So as you can see, this is our postal code and it comes He's under Dimensions. Why because it is currently. Number as we can see, it's a whole number and what we gonna do. We're going to edit this Str. String and put in postal code and okay.

Now as you can see the type conversion, which is basically the same thing as the postal code has a string data type. The postal codes will remain the same, but their data type is going to change that is what the string function does. And with that we come to the end of type conversion functions. Let's move on to aggregate functions. There are a few more than the type functions here. So basically what aggregate functions let you do is That they allow you to summarize or change the granularity of your data. For example, if you want to know exactly how many orders your store had for a particular year you can use the count D function to summarize the exact number of orders your company had and then break the visualization down by ear.

So there are a bunch of aggregation function out here. Most of which we have already used as number functions before and if we haven't used them as number functions before for while we have dragged it to our marks panel. Most of you have seen how they work. So let's start with the first one. The first one is attribute. You might have seen through the course of this video how I have changed the default aggregate from some to attribute many times over in this video itself.

It basically Returns the value of the expression like it has a single value for all the rows. So it is like one standard value the null values, hence. Ignored next as the name suggests. This is the average expression. It Returns the average of all the values in the expression since I haven't used it before I am going to show how this works. So suppose I bring in my sales here as you can see the default measure is some I can break it down to the average and it shows average sales nothing new.

No rocket science next we Have correlation coefficient which basically Returns the Pearson's correlation coefficient for two expressions. Now the Pearson correlation measures the linear relationship between two variables and results range from minus 1 to plus 1 where 1 denotes an exact positive linear relationship as when a positive change in one variable implies a positive change of the corresponding magnitude in the other now 0 denotes no linear relationship between the And negative 1 is the exact negative of a relationship. Now. What I'm going to do is I'm going to use a table scoped level of the detail expression. So I'm going to go ahead and And this is what it gives it's between negative 1 to positive 1 and this is how far the relationship is linear. So you can say that the relationship between profit and sales is fairly linear because it's on the positive side and with a level of detail expression the correlation you can run all over the rose and the view would show the correlation of each individual point in a scatter plot if you want to do that.

Next we have count which Returns the number of items in a group null values are obviously not counted and then we have count D which Returns the number of distinct items in a group which basically means that if a certain item is there twice in a certain group count D is going to count it as one item. Then we have the covariance expression. This basically returns a sample covariance of two expressions. It's basically quantifies how two variables change together a positive covariance indicates that the variable tend to move in the same direction. When a larger value of one of the variables tend to correspond to a larger value of the other variable on an average. Now the sample covariance is the appropriate Choice when data is random and it is being used to estimate the covariance for a larger population. Like we found out the correlation between sales and profit we can also find out the covariation between sales and What next we have Co variance of the population it basically does the same thing as covariance.

But this time of the entire population next. We have Max which we have tried Thrice before it Returns the maximum expression across all records. We have median which Returns the median expression across all records. And then we have the main which Returns the minimum expression across the records next. We have percentile what it does is it returns? The percentile value from the given expression corresponding to the specified number now the number must be between 0 and 1 obviously and this function is available in Microsoft Excel and text file connections Google analytics or Salesforce Cloudera Hive Oracle 10 and deleted data resources of Oracle exist solution 4.2 and the later versions and sybase. Next we have standard deviation which Returns the statistical standard deviation of all values. In the given expression based on a sample of population now the same thing for the entire population and not the sample is the standard deviation of population it basically Returns the statistical standard deviation of all values in the given expression based on a biased population. Next you have some which is also the default aggregate which is allotted to all the measures and dimensions that you use it Returns the sum of all, All values in an expression and of course it can be used only with numeric fields and null values will be ignored then you have variance which Returns the statistical variance of all values in the given expression based on the sample of your population.

And finally we have the population variance which gives the statistical variance of all the values based on an entire population and not just a sample. We're finally at our last segment logical functions. Now why should you be using logical functions now? This is basically used to calculate Boolean logic. If you don't know what that is, it basically means you give a certain condition and it tells you whether it's true or false there is no other output than true or false. Now most of these things.

We've already seen through the many functions that we have covered today and the rest are very common logical functions if you are from any coding background, so what I'm going to do For this particular segment is that I'm going to go through all of these functions first and then go to my Tableau desktop to show you a tiny demo. So first of all and function now basically this performs a logical conjunction of the two functions. So basically it's the logical and if you look at the given example, it says if your Market is equals to Africa and the sum equals sales both are greater than emerging threshold, then it should give the output while performing Notice how both of these conditions have to be true for Tableau to give you an output of well-performing that is what a logical and does for logical and to work. Both of the conditions need to be true. Next you have case if anybody of you are from the coding background, you can relate it with the switch case. It performs a number of logical tests and returns appropriate values the case functions evaluates Expressions Compares it to a sequence of values.

And returns a result when a value that matches the expression is encountered case Returns the corresponding return value. If no match is found then the default:return expression is used and if there is no default return and no values match, the null is returned case is often easier to use then if else and statements like that. Typically if you use an IF function to perform a sequence of arbitrary test, you can use a case function to search for a match. When expression but a case function can always be re-written as an IF function. Although the case function will generally be more concise. So many times you can use a group to get the same results as a complicated case function. We learn more about if in the later parts of this logical functions segment next we have else now. This is the counterpart of if so, if something then a condition has to be either Old or not fulfilled it will give you an answer else.

It will give you another answer. It's as simple as that, it tests a series of Expressions returning the den value for the first true expression. Then you have else if if you have a number of if statements before your default else, then you use else if it tests a series of Expressions returning the then value for the first row expression again, then you have end which as the name suggests. It marks the end of an expression. Then you have if as we have already discussed if basically tests in expression and then returns a corresponding value, then we have if null which Returns the expression one.

If not null otherwise returns expression to it is basically like a regular IF function except for it just checks for one condition. If the expression you've put in is null or not. Next is I if which checks weather condition is met and turns one value if true another value if false it's basically like a test case and an optional third value or null if unknown next we have is date which returns true if a given string is a valid date.

It's no big deal. Then we have is null which returns true if the expression does not contain valid data, which is if your expression is null then we have Max which we have discussed many times before it returned. Maximum of all the records or maximum of two expressions for each record. Then we have minimum which does just the opposite.

It Returns the minimum of all records or the minimum of two records. Then you have the logical not which performs a logical negation on an Expression. So basically if the field you have put in does not fulfill the condition then it gives you the corresponding result. Then you have the logical or which basically Means that if you have put into conditions, either or of them have to be true, then it will give you a result. Then we have then which we have discussed before if a certain field fulfills a condition the corresponding then part is then fulfilled now here we have a venn function which finds the first value that matches the expression and Returns the corresponding return.

It's kind of like if and then case except for there is no if Then you have ZN which returns your expression if it is not 0 or null we have discussed this before now, let's go to our Tableau desktop and create a logical calculation to see how you can use this. So now again come back to our Tableau desktop one last time for this session.

So again, I am from the data pane drag state to the Rose shelf as you can. In see a table, like this should ideally appear then I am going to take category place it at the Roses. Well, this is nice. I'm going to analysis as I've done many times before in this session create calculated field here. I'm going to name this kpi, which is key performance index basically shows you how your company is doing internally and externally internally azing different departments and externally as in in the market going to do is see some profit we can see this here. Now this calculation quickly checks. If a member is greater than 0 if so, it will return true.

If not, it will return false. It's kind of like the if and then statement so when finished you can go ahead and click OK as you can see this new calculated field appears under the measures and the data pin just like your other fields you can use it on this visualization. So I'm going to drag this to the color on the mocs card. animal turn this whole thing into so I'm going to take the sales and put it in the columns. You can now see which categories are losing money in the states. So all these orange ones are the ones in profit and the blue ones are the ones at loss.

And that's the simplest use of The Logical functions. You can use this for way more because Tableau is capable of not just pretty graphs but a good drill down of data with that. I am going to close the session Tableau was made with an intent to analyze data in a much easier and efficient way than we were doing all these years with these traditional methods. But if you have to stop thinking about how to use the tool to solve a problem the state of flow is broken one common cause of this is the need to work with data that has been aggregated to different levels of detail. Hi. I'm a pastor from Eddie Rekha. And today we're going to talk about levels of detail in Tableau before we begin. Let's look at the agenda for today.

First up. We are going to talk a little bit about what aloni actually is what it does then we'll talk about the various calculation in LOD which are The include the exclude and the fixed calculation then we can talk a little bit about aggregation and LOD Expressions followed by nesting and inheritance in LOD. Then we're going to talk about the data sources supported by level of detail and Tableau. Finally. We have a short demo on how to create some simple LOD expressions in Tableau. Then we're going to talk about table calculations in LOD and finally discuss a few limitations of Valerie. So without Much Ado, let's get straight into the module now the Few questions, which always arise while dealing with data that has been aggregated to intricate levels of details and these questions are often simple to ask but really hard to answer the sound something like can I plot the number of days per quarter where my company had more than a hundred orders? How can I find the biggest deal each salesperson has ever closed then show the averages by manager.

How can I tag every customer by the year he or she first became a customer and then use that tag to group The Sales, in order to address these types of questions Tableau 9.0 onwards introduced a new syntax called the level of detail. Now this new syntax both simplifies and extends tableaus calculation Language by making it possible to address level of detailed questions directly. So in simple terms level of detail Expressions represent an elegant and Powerful way to answer questions involving multiple levels of granularity in a single. Sure now granularity and aggregations in LOD or inversely proportional to each other how it works is the more the level of granularity. The less is the level of aggregation and vice a versa. So here are the shelves which affect your LOD Aggregates you have your columns rows pretty much everything except for your pages and filter shelf and these are the shelves which do not affect your LOD Aggregates.

These are your pages filters. And tooltip now all these LOD expressions are segregated into three types. We have included which basically calculates at a lower level of detail. We have fixed LOD expressions with specify the exact level of detail and we have the exclude level of detail Expressions, which calculate at a higher level of detail. So first, let's talk a little bit about how you work a problem in level of detail. Now this map shows the restaurant inspection. Doll from Yelp in the greater Edinburgh area. So the data in the view is aggregated based on the LOD which in this case consists of city and state and is more aggregated than the underlying data source. So the selected point in the image shows the average user fans for all the restaurants in Newbridge Edinburgh adding more granular Dimensions to the view will result in a less aggregated LOD. For instance. We could add business ID to the visualization by popping it on the detail shelf to see the average user fans for each individual business by doing so will also change the visualization every single business will appear as a circle on but what if we don't want the visualization to change what if we want to determine the total user fans for each business ID and average those values for each City and finally show only one Circle per City.

What we want to see is the average number of fans per restaurant in each. Now this will require adding an additional Dimension to the view without dragging that Dimension into the visualization a level of detail expression will allow us to do this. Now this expression tells Tableau to perform the aggregation for each business ID regardless of other dimensions used in the LOD, you can use this expression to calculate the total user fans / business ID after dragging this new field into the view.

We can then average those values per setting now the fans / business field have been added to the color shelf as you can see on your left now new bridge has the highest average fans per business with a hundred and eighty five pans a value that was computed using the fixed LOD expression by using the fixed operator in our LOD expression. We gain insight into which cities have on average more fans per business ID meaning those cities with a darker shade of blue have more popular restaurants, or maybe even the city could More residents and hence more total fans / restaurant. Now what you see up top is how the level of detail expression is structured. You have your scoping keyword and then you have the dimension declaration and finally the aggregate expression. So you're scoping keyword as in your include exclude or fixed your dimension declaration in this case your business ID and your aggregate expression is what you want to do with that declaration like Here you are going to use some expression here is the structure given more clearly for the level of detail expression.

Now as a result of that fans per business field have been added to the color shelf and it shows that Newbridge has the highest as computed using an LED expression average fans were business with a hundred and eighty-five fans. Now, let's talk a little bit about the include calculation. So the include LOD expression will add Dimension to the LOD then Keyboard creates an expression that is less aggregated, which also means it is more granular than the LOD the specified dimension of first added to the LOD before the calculations are performed. Now notice that the include expression is used in The View as an aggregated measure. In fact, all include expressions are either used as measures or aggregated measures when placed on The View, let's talk a little bit about the exclude calculation now exclude calculation. Basically means calculating at a higher level of detail using an exclude keyword will exclude the desired Dimensions from the calculation Tableau first removes the excluded Dimension from the LOD and then performs a calculation as of the dimension was not present at all.

The result is then displayed visually the graphical representation of how Tableau performs an exclude LOD expression is depicted in the diagram that you can see next. Let's talk about the fixed calculation now LOD Expressions also. Oh The door to creating an aggregation level completely independent of the LOD something that was previously only possible by the custom structured query language presenting to you. We have the fixed calculation which basically specifies the exact level of detail this LOD expression will fix the level of detail to each Dimension and does the aggregation specified in the calculated field regardless of any dimension in the view if you look at The sequence of filters and Tableau fixed calculations are applied before Dimension filters. So unless you promote the fields on your filter shelf to improve view performance with context filters. They will be ignored include and exclude level of detail expressions are considered after Dimension filters. So if you want filters to apply to your fixed level of detail expression, but don't want to use context filters consider rewriting them as include or exclude now, let's go ahead and look at aggregations.

I'm level of detail now the level of detail of the view determines the number of marks in your view. So when you add a level of detail expression to the view tab low must reconcile two levels of detail the one in the view and the one in your expression the behavior of a level of detail expression in the view varies, depending on whether the Expressions level of detail is coarser finer or on the same level as the level of detail in the view. So what do we mean by Corsa or fine and this case so if I said the level of detail Action is coarser than the view level of detail.

I mean that it references a subset of the dimensions of the view. For example for view that contain the dimensions category and segment you could create a level of detail that uses only one of these Dimensions. So when I use this expression here, the expression has a coarser level of detail than the view it bases its values on one dimension that is segment. Whereas the view is basing its view on two dimensions. I'm category. The result is that using the level of detail expression in the view causes certain values to be replicated. That is they will appear multiple times. Now if I say the level of detail expression is finer than the view level of detail. It references a super set of Dimensions In The View when you use such an expression in the view Tableau will aggregate results up to the view level. For example, the following level of detail expression references two Dimensions when this occurs Expression is used in a view that has only segment as ass level of detail.

The values must be aggregated here is what you would see if you drag the expression to a shelf and aggregation in this case average is automatically assigned by Tableau, but you can always change the aggregation. If you need it now adding an LOD expression to the view whether level of detail expression is aggregated or replicated in the view is determined by the expression type which is fixed. Execute and whether the Expressions granularity is coarser or finer than the views now for include the LOD will have either the same level of detail as the view or a finer level of detail than the view.

Therefore values will never be replicated for fixed level of detail Expressions can have a finer level of detail than the view a courser level of detail or the same level the need to aggregate the results of a fixed level of detail depends on what dimensions are in The View. And finally for exclude level of detail Expressions replicated values will always appear in the view as we had discussed before when calculations include the exclude level of detail Tableau defaults to the attr aggregation to indicate that the expression is not actually being aggregated and that changing the aggregation will have no effect on The View now, let's discuss a little nesting in the level of detail now Tableau does not limit you to write single simple.

You can Nest as many Expressions as you want according to your requirements. So when you approach this kind of a problem, you have to understand a few rules that come with this inheritance property. So there are two types of inheritance in table calculation. One is the impact of fixed expression, which we had earlier mentioned has impacts on where it is evaluated and which filters affected and then there is the dimensionality itself. So in this case, if you look at the first one I am doing a fix State and nested calculation I'm saying include customer. So include as you all know in head is from its surroundings, if you build the dragging and dropping by itself, it will inherit from the parent calculation. So it will include the state it will do the same things as writing State customer instead of it and it won't be impacted by filters because the parent is a fixed calculation with that. Let's move on to the data sources that are supported by level of detail. So here I've made a list on Data sources and whether they are supported or not supported by level of detail.

Now, let's move on to a Hands-On of how to create level of detail expressions in Tableau. Now, the question is how to create these LOD expressions for that. We will have to move on to our Tableau desktop. So let's head there now for your LOD expression first up prerequisite is that you need to have a visualization already set up. I've created a very simple bar chart over here with three. Agents from the dimensions data pane in the columns and sales in the Rose shelf now Step 2 instead of sum of all sales per region, which is given by Tableau by default this one right here. Perhaps you would also want to see the average sales per customer for each region, but this you can use an LED expression. So I'm going to go to analysis and create a calculated field.

In this calculation editor that opens I'm going to name this sales per customer and then I'm going to put in this expression. So include customer name. And then I'm going to apply it to my visualization and click. Ok. So the newly created LOD expression is added to the data pane as you can see. It's right here sales per customer under the measures pain. Now we are going to use this LOD expression in the visualization. So from the data pane under measures, I'm going to drag this sales per customer to the Rose shelf and place it to the left of some of sales see right here on the same, Rochelle. I'm going to right click on sales per customer and select the measure some and then take the average you can now see both the sum of all sales and the average sales per customer for each region.

For example, you can see in the central region that the sales totaled approximately $500,000 with an average sale for each customer being approximately 800 US Dollars. Now this was all about the include expression. We're going to do the same for exclude and fixed. now for exclude expression, we're basically going to try and exclude region from a calculation of some of sales and to illustrate how this expression might be useful first consider the view that you see on your screens right now, which breaks out the sum of sales by region and by month, so I'm going to be creating this expression the same way I did before You see a new measure being created here called exclude region. Now, we are going to drop exclude region this measure that I just showed you on color Shades, which is right here. We basically showing the total sales by month, but without the regional component, I'm going to change the colors a little bit.

So it's more prominent. So let's go with this orange and blue Divergent color. All right. So this is how the exclude expression is created. Now, let's move on to the fixed expression, which is another sheet. Now as I had mentioned before a fixed level of detail expression compute some value using specified Dimensions without reference to the dimensions of the view. So the fixed LOD expression. I'm going to use now computes the sum of sales per region. So we're going to do the same thing, which is go to analysis and create a calculated field. As you can see this measure has been created here. So what I'm going to do right now is I'm going to take this measure and please sit on the text to show the total sales per region. So the view level of detail is region plus state but because fixed level of detail Expressions do not consider the view level of detail, the calculation only uses the dimension that is referenced which in this case is region because of this you can see that the values Individual states in each region are identical with that.

Let's switch back to our presentation. This is a very common question of how the level of detail compares to the table calculation. And now that we have level of detail do we actually need the table calculations? The answer is yes, you will still be needing table calculations, but the LOD is here to take care of a lot more of the things now. First of all table calculations are generated by query results. Whereas the LOD expressions. Are generated as a part of query to underlying data source table calculations can produce results either equal to or less granular than said LOD Dimensions that control the operations of a table are separate from calculation syntax in table calculations. Whereas in LOD Expressions Dimensions that control the operations of analog expression are embedded in the expression itself. Another big difference is table calculation can be used as aggregated measures and The Expressions can be doubled for various other constructs the filters on the table calculation act as a hide whereas filters on the LOD act as an exclude.

And finally, let's discuss the limitations of level of detail. Now the a few limitations and constraints that apply for the level of detail Expressions. Now level of detail Expressions that reference floating-point measures can behave unreliably when used in a view that requires comparison of the values in the expression the LOD expressions are not shown on the data source page while referencing a parameter in a dimensionality declaration. We cannot use parameter values. We have to always use the parameter name. So if you do not know the parameter name, then you going to face a problem with dimensionality declaration and finally would data blending the linking field from the primary data source must be in view before you can use a level of detail expression from the I can read data source, in addition. Some data sources have complexity limits and Tableau will not disable calculations for these databases.

But query errors are a possibility if calculations become too complex with that being said LOD expressions are a powerful new capability of Tableau, which allow us to easily solve problems that previously required very complicated formula. They allow us to intuitively define the scope of calculations and stay in the state of flow as we Our data they are not a new form of table calculations, even if they can replace many of them, but they do open doors to new possibilities contrary to popular belief elodie's and table calculations operate very differently from each other.

And finally, I'd like to say that LOD Expressions represent a vital step towards the goal of complete flow where all questions are simple and elegant to answer. It has been well established that Tableau is not just meant for pretty visualizations. It is also helpful in intricate calculations aggregate functions and many more drill down procedures. Hi all this is a pasta from Eddie Rekha. And today we are going to talk about one such feature in Tableau called a parameter. So we're going to start out by discussing what our parameters in Tableau then we are going to move out to our demo machine, which is our tableau. Desktop we are going to start there by connecting to our data sources.

Then we are going to create a parameter in Tableau. Then we are going to use the parameter in a calculation learn a little bit about parameter control. And finally we're going to use our parameter in our visualization and see how it affects our data now before I go much further. Let me request you all to go ahead and hit that subscribe button. So you never miss a new weekly. Deal from your favorite tech Channel at Eureka YouTube channel. So without Much Ado, let's get started. So what exactly is a parameter in Tableau think about it like this. What if you need a component for your visualization that is not exactly in your data set parameters and Tableau will allow you to provide that value which you're going to pass to Tableau.

Now this particular feature will allow you to use aggregated. Use that aren't readily available in your data set. It will help you incorporate these values into your dashboards and reports directly. Now after creation and users can control the import to see the results of the effect of the parameter easy, isn't it? So what exactly is a parameter now any value that is passed to your program in order to customize it for a specific purpose is called a parameter now, it could be anything. Say a string of text a range of values or any amount in rupees or dollars just to name a few parameters will help you experiment with some what if scenarios suppose you are unsure which feels to include in your view and which layout to not what layout would work best with your viewers giving them the choice you can incorporate parameters into your views your charts your graphs and your calculations to let your viewers.

Is choose how they want to look at your data. Now when you use parameters it is of utmost importance that you need to tie them to the view in some way one way to do. This is why our calculations you can use calculated fields, which are Incorporated in your visualizations in Tableau second. You can display the parameter control in the view for your users to select from the parameter now finally, you can reference parameters in parameter actions. Which basically means you can use them in your graphs and see the effect it has on your data. So now that you know a little bit about what parameters are just theoretically knowing about this concept wouldn't obviously do you much good so the next few segments I shall carefully guide you through the process of creating and using these parameters in Tableau from this point onwards.

We are mostly going to be on our demo machine, which is Tableau desktop. So let's get Started so when you open your Tableau, it kind of looks like what you're seeing on my screen right. Now. What I'm going to do is I'm going to connect to the sample Superstore that is already provided by Tableau. Now why I'm doing this is so that you guys do not have a difficulty in finding the data set that are using you can directly go connect here and follow along. Now. This is what my data set looks like.

I have the sample Superstar, which is basically a collection of many stores spread. Across the United States, it gives you the country city and state the customer name your shipping details your order details your categories and subcategories of your products. Basically all your sale information sales discounts profits profit ratio. So on and so forth. This is what your data looks like. Now, let's move onto our sheet. So by no means am I my going to give you like a full beginning with Tableau tutorial. We already have a few of those kind and we have one coming up pretty soon and updated one so you can go ahead and look at that in our playlist. So basically what I'm going to start out doing is I'm going to create a basic graph. So let's see. I want to create a sales according to the order date sort of a graph. It's going to be a line graph.

I'm going to put the order date and my columns and there is a measure called sales that I'm going to put on my rose shelf. It gives me a graph like this, but this isn't very informative as we just have four years now. I want a more elaborate graph. So I'm going to go ahead and click on the spill with says ear. I'm going to go to the options. Select more and then go to custom. Okay. Now this is the custom date dialog box as you can see it only has the auctioneer selected.

I'm going to go down and I'm going to go select month / here. So it's going to show me all the months through these four years. All right. So now as you can see, we have all the months from 2015 through 2018. We have a much more elaborate graph something more. Walk with you shall be easily end up with a graph along the lines of this. It should look something like this. Now. We are going to be creating a parameter in Tableau. So basically the scenario I am trying to create is a what if scenario like I had mentioned before so I'm going to say for example, what if the sales has been hiked up by 3% Now this detail is not given readily to me in my data set.

Set so I obviously have to create a parameter for it. So basically this is a parameter which I'm going to be using in a calculated field to create a calculated field you go to analysis and create calculated field or alternatively you could also go to this down arrow key near dimensions. And the first option you get is a calculated field now before we make the calculated field, let us create the Our meter that we are going to be using in the calculated field. So the second option from that was create parameter, so I have my dialog box. So I'm going to go ahead and name this like if because it's an if if scenario if sales parameter, I'm going to move the data type to integer from the drop-down menu current value.

I am going to keep as zero trust me. There's a reason why I'm doing this just pray for matters. Is automatic now I'm going to go ahead and select the range minimum. I'm going to be keeping 0 maximum the default as hundred and step size. I am going to keep as to so with that. I have created my parameter. It's going to be an integer type parameter ranging from zero to hundred now, as you can see in the bottom here in the parameters set you can see and if sales parameter the one that we have just created. Remember that I had told you that you have to use your parameter and tight to your view in some way. And the first way that I have told you was to use your parameter in your calculation. And that is exactly what we are going to do.

Now in the scenario. We want to use our parameter and a few Tableau functions to create a calculated field to add to our graph and then we are going to see its effect on our data. So we're going to go ahead and and create a calculated field now. I'm going to name it the same as my parameter. Except for I'm going to name it calculation and here I am going to be throwing in a formula which if you want to know more about you can go ahead and check a tutorial that we have on functions in Tableau. So you understand what these functions do basically So I'm putting in this formula here and at the bottom it says calculation is valid now. I cannot stress enough. So I'm going to say it one more time in this video. I've mentioned it many times in many videos and lives before that. This is something which shows that Tableau is a really really smart software if suppose. I remove a parentheses from here in the bottom.

It is going to show the calculation contains errors, and if you hit the arrow button here it's going to tell you. What error it is this way? It prevents you from making mistakes right from the get-go. You do not have to go much further in your process. This is going to be a small process for demo purposes. But usually Tableau is used in an industry level and there once you have gone much further in your procedure coming back and correcting mistakes will be Troublesome. So I'm going to hit okay and in your measure section you can See your if sales calculation is here. Now. What I want you to do is notice how your calculation the parameter that we created is going to interact with your sales measure in the segments that follow next. What we're going to do is parameter control now coming back to the Tableau main menu, as I had just mentioned you can see your calculation field in the measures pain and your parameter in Parameter spin this is your data window. So I'm going to click on this and click on the option show parameter control on the right.

You see this particular option, which is your if sales parameter, this is your parameter control currently it is there in this slider form giving you your range is zero to a hundred which we had selected. You can always go ahead and change it from Slider form to your type and format. ERM I prefer the slider form over the type and form it's easier to operate for me. You can go ahead and make a choice. This is the top right of your view.

And this is where by default your parameter control filter is always displayed now, I'm not going to show you how it is used right now in the next segment. You will see what its use is. So finally we are going to be using Tableau parameters in our visualization. This is the part most of you might already be waiting for. Okay. So starting out. What I want to do is I want to switch this for measure values and I am only going to be keeping. my f sales and the regular sales which we had made the graph with before. Okay. Now I have my calculated field if sales right here. And so I have my sales also right here. Okay now because on your right you can see your parameter control is at 0 you might be able to see just one graph. But as we start moving the slider up suppose I keep mine at 30 you instantly get two of your graphs.

One of them is your if calculation your if sales calculation, which is your calculation grown by 3% your sales grown by 3% Right and the other one is your regular sales. You see the difference once you click on your parameter control and set it to any number which is visible. Like I would suggest go above 10. I have set it as 30 you can see your dual axis graph. You can go ahead and change the color if you want to but I'd like to keep them in blue itself because I think it's pretty visible. But you can always go ahead and change the colors. If you like. Now these lines represent the running values of sales from your data set and the calculated sales simultaneously and you successfully Incorporated your parameter in your visualization along with your control that you have on your right and that is how you create and use parameters in Tableau now parameters are Dynamic and useful elements for You to add interactivity and flexibility to your dashboards and reports. It is a very versatile tool and can be used in way more than what I showed you in this demo. It can be used in various calculations sets equally well now this is one of the many smart features that are there in Tableau, which is emerging as one of the hottest Trends in business intelligence in 2019.

And also if I might add it is one of the easiest First Data visualization tools to learn one of the most interactive and smart software's and if you look at Google Trends, it seems like there can be no better time than right now to get certified in Tableau to start learning Tableau in a world with generates and consumes 2.5 quintillion bytes of data a day.

Organizations are bound to look for new methods to And combine data in order to obtain Optimum efficiency one such method of combining data is data blending in Tableau. Hi all this is a pastor from Ed Eureka. And in this module, we're going to talk all about data blending. But before we begin, let's discuss our agenda for today. So first of all, we're going to talk a little bit about the objective of data blending then we're going to talk about what data blending essentially is and how it works in. Whoa, then you're going to discuss a concept called joining and see how is it different from data blending. Then we're going to see how can you do this? It's going to be a very short demo a few simple steps. And finally we're going to discuss a few limitations in this process. So without Much Ado, let's get straight to the module. So what is the objective of data blending in Tableau? Why do we need data blending now? Let's suppose you have transactional data stored in Salesforce.

I'm quarter data stored in an Excel workbook the data you want to combine stored in different databases and the granularity of the data captured in each table is different. So in such a case, you use data blending now data blending could be very useful under a few conditions. Like you want to combine data from different databases that are not supported by cross database joints now cross database joins. Do not support connections. Two cubes take Oracle essbase for instance or some extract only connections take Google analytics as your example in this case set up individual data sources for the data you want to analyze and then use data blending to combine the data sources on a single sheet.

Next is when you have data at different levels of detail. Now sometimes one data set captures data using greater or lesser granularity than the other data. For example suppose you are analyzing transactional data and quota data. Now your transactional data might capture all transactions. However, what our data might aggregate transactions at a quarter level because the transactional values are captured at a different level of detail in each data set you should use data blending to combine data now third case is when you have a lot of data typically joins are recommended for combining data from the same database. So join is basically another method of data merging in Tableau. We shall discuss it in depth in the later parts of this module now joins a handled by the database which allows joints to leverage some of the databases native capabilities. However, if you're working with large sets joins can put a strain on the database and significantly affect performance in such a case data blending might be of great use to you because Tableau handles combining the data after the data is a aggregated there is less data to combine when there is less data to combine generally performance improves.

And finally you can use data blending when your data needs some cleaning if your tables do not match up with each other correctly after join setup data sources for each table make any necessary customizations, which basically will include renaming columns changing column data types creating groups and so on and so forth, then you can use data blending together. Combine the data now that we know when to use data blending, let's find out what data blending actually means. So data blending is a method to combine data that supplements a table of data from one data source to another data source for people who use SQL it is basically an advanced version of your left. Join now, what is a join and how is it different from blending data in Tableau now data blending skin? Emulates a traditional left join which I had mentioned a few seconds ago.

The main difference between the two is when the join is performed with respect to aggregation. Now when you use a left join to combine data a query is sent to the database where the join is performed using a left join returns all rows from the left table and any Rose from the right table that has a corresponding row match in the left table. The results are then sent to Tableau to be aggregated. For example suppose you have the following tables if the common columns are user ID a left join takes all the data from the left table as well as all the data from the right table because each row now has a corresponding row to match.

Now, how is it different from data blending now, when you use data blending to combine data a query is sent to the database for each data source that you're using the results of these queries including the aggregated data. A sent back to and combined by Tableau now take for instance. You have the following tables again the same tables if the linking Fields again our user ID on each table blending your data takes all the data from the left table and supplements the left table with the data from the right table The View uses all the rows from the primary data source the left table and the aggregated rose from the secondary data source, which is the right.

Able and it is done based on the dimension of the linking Fields. If there are multiple values for Rose and asterisk is shown measure values are aggregated based on how the field is aggregated and the view in this case not all values can be a part of the resulting table because of two reasons. First a row in the left table does not have a corresponding row match in the right table as indicated by the null value and second there are multiple Values in the rows in the right table as indicated by the asterisk or the star sign now suppose you have the same tables as before but the secondary data source contains a new field called finds again.

If the linking feels our user ID blending, your data takes all the data from the left table and supplements it with the data from the right table. In this case. You see the same null value and In the previous example in addition to two things now because the fines field is a measure you see the row values for the finds field aggregated before the data in the right table is combined as for the previous example a row in the left table does not have a corresponding Row for the fines and that is why it is indicated by the second null value. Now, how can you blend your data now? You can use data blending when you have data in separate data sources that you want to analyze together on a single Sheet example.

I'm going to show you now demonstrates how to blend your data from two different sources now for this I'll be moving on to my Tableau desktop and here I'm going to be using two data sources name the sample Superstore, which is already included in the sample data sets of Tableau and the sample coffee chain, which is another very easily available data set for Tableau online.

So first, I have already loaded the sample coffee chain to Tableau and now here is its metadata. We see profit margin sales cogs total expenses marketing inventory budget profit. Margin Budget Sales, etc. Etc. And this is all in an MS access database file here. You can see all the various tables and joints that are there in this query right here next step. Is adding a secondary data source. So what we're going to do is we're going to add a secondary data source named Sample Superstore by again following the steps. It's pretty simple. Actually. All you have to do is click on this add button and you will find the data set right there. You can search for the data set here. And that's it here. We have both of our data sets now blending your data. These are both our metadata has we're going to go to our Sheet now what we can do is we can integrate the data from both of the sources based on a common Dimension. So when I select this state, so what I'm going to do, I'm going to select the sample Superstore go to the profit ratio put it in my columns, then I'm going to Rose.

I'm going to be selecting state. Putting it in the row shelf. Then I'm going to select this chart. Let's try the Gantt bar. Nope automatic it is. Now if I go to my coffee chain query and if I look at my state Dimension, what do I see here? This is a small chain like image that is appearing near the state Dimension. This basically indicates that the common Dimension between the two data source is something called the state if I open my other data sources as well, which is the sample Superstore. This is my common Dimension and the chart here basically shows Shows how the profit ratio varies from each state in both the superstore and the coffee chain shops. And that was it for our Tableau desktop. Let me go back to my presentation where we can go ahead now limitations of data blending. Now, what are the constraints that apply to this method? First of all blending with non-additive Aggregates now, there are some blending limitations around the non-additive Aggregates such as Count the median and raw SQL aggregate when you blend on a field with a high level of granularity suppose date instead of your let 's say queries can be slow down.

So basically the speed of the query gets compromised now values appear after blending the data sources now null values can sometimes appear in place of the data you want in The View when you're using data blending and this happens because of a few reasons, it can be so that the second data source does not contain values corresponding to the primary data source, or the data types of the fields. You are blending are on different levels of detail or the value in the primary and secondary data sources use different casing it can be anything but the null values sometimes after data blending appear in place of the data you want to view and finally sorting by feels is unavailable for data Blended measures, but despite that data blending.

Is a whole new approach to merging of your data. It saves you a lot more time makes your system way more efficient and optimizes the data cycle as a whole a tableau developer today is one of the most sought-after job roles in the bi industry. So what does it take to become a tableau developer? Well, you have all come to the right place. Hi all I'm a pastor from Eddie. Erica and in this module, we are going to talk all things career when it comes to Tableau.

But before we begin, let's talk a little bit about our agenda for today here first. We will be talking a little bit about Tableau followed by the role of a tableau developer. Then we shall discuss the responsibility and job profile of Tableau developer later. We shall explore the required skills and abilities for the same job role. And finally we're going to talk a little A bit about getting certified in Tableau and improving other technical skills. So without Much Ado, let's get straight into the module. So what is Tableau now tableaus a platform that focuses on understanding data and uses its potential in business strategy. It's a platform that comprises of creating dashboard reports visualizations and deriving insights and feedback to improve on larger systems. The next logical question is who is eight. Tableau developer now a tableau developer create solutions for data visualization to enhance the business processes. This job comprises of various tasks such as working with developers creating Tableau reports creating bi visualization and participating in feedback sessions to improve systems.

Now this job is perfect for people who work well as a part of a team and who have problem solving skills that can manage their time productively to meet deadlines. Adaptive developer is usually someone that is proficient in data visualization mathematical reasoning database skills and extract transform and load increase s's pursuing a career as a tableau developer can mean many things a few of those. I want to discuss with you today. First of all the connectivity option you get the reason why Tableau stands out amongst all the bi tools is that there is a wide range of connectivity. See options now Tableau can connect to any data. You can possibly think about starting from spreadsheets to databases and even big data you can access warehouses Cloud applications like Salesforce and even connect to Cloud database like Amazon redshift. It has a web data connector and it is used to pull API directly from web in order to connect to any desired data source. Now, let's talk a little bit about pursuing a career as Tableau developer Tableau is known as the leader in bi tools and it has been crowned the best by the it research giant Gartner Gartner's magic quadrant mentioned Tableau for the fifth time in a row as the best amongst a blos competitors like Microsoft sap and click in addition to having a great demand for Tableau experts.

There are always rewards to offer if you browse through the job portals like indeed and AngelList. You can find plenty of job postings Tableau professionals. Get the best of salaries in the mighty come bi industry with an average of 91 thousand dollars per annum. There are tons of jobs available, which require Tableau as images skill set. Now, let's take a look at the responsibilities that come with this lucrative job profile now Tableau developers responsibilities vary depending on the type of organization they work for Here are a few job descriptions that I have picked out by major companies for a tableau developer. This is the one by cognizant which says they want somebody with industry experience with Hanson in design and development of Tableau visualization solutions.

They want somebody with creation of users groups projects workbooks and appropriate permission sets for Tableau server log ons and Security checks apart from this. They also want the Channel to have strong data warehousing and business intelligence skills. Next we have a job description by Bosh. They've kept it pretty simple. They want an engineering graduate with at least two to three years of experience on Tableau minimum a year or two of experience on Tableau professional or analyst software suite experience and desktop and server architecture creation configuration and deployment of Tableau servers in visualization. Jan and Publishing authorization Concepts along with some good communication and analytical skills apart from which they want somebody with experience in working with multiple data sources and handling large volumes of data. Next. We have a job profile by Tech Mahindra. They have kept it pretty simple as well. They want somebody with a strong understanding of advanced Tableau features including calculated Fields parameters table calculations Joy. coins and dashboard action the shoes they expect you to fill generates Tableau dashboards with quick context Global filters parameters and calculated fields on Tableau 10 point x reports apart from which from what I see they need somebody with strong structured query language skills in building complex queries triggers indexes involving multiple tables from different database schemas, but all of these job descriptions basically boil, Don't do these major points a tableau developer is responsible for creating Technical Solutions, which basically means the primary objective of this developer is to create a solution which matches the needs of the business.

This can be done by finding Innovative resolutions and translating their requirements. Another responsibility is working with data storage tools, which preserve data within organizations. This is also known as online analytical processing. A processing or olap apart from which they need to conduct tests for which they develop database queries and conduct unit test to troubleshoot and analyze the issues that arise this process is an ongoing part of the development that occurs continuously throughout the project. They also need to enhance systems, which is a crucial part of the job. It means they have to evaluate and improve existing systems. This also includes collaborating with other teams within the business. To incorporate new systems to streamline company process and workflow. They also need to create technical documentation for completed projects to communicate with senior staff members and colleagues within the organization for reference. And finally they need to use bi Technologies structured query language data analysis and ETL tools for storytelling and forecasting of data.

Tableau represents data, like no other tool with unique features like forecasting and storytelling one can even connect to the data personally and understand the depth of the analysis having said that let's move on to look at certain skills that are required to fulfill these responsibilities of a tableau developer. Now the required abilities to become a tableau developer are as follows a tableau developer should have a bachelor's degree in business.

Computer science or any similar field they require experience in the whole life cycle development of applications at an Enterprise level. They need to have proficiency with structured query language has a large data sets. They should have excellent analytical skills as they are needed to analyze the requirements of a client or a business. This role also demands to work with software from the beginning till the end of the project. So they need to solve any issue that occurs during the development. Onstage apart from that this profile requires the ability to create innovative solutions to problems with in business. They need to be self-motivated for finding Solutions and improvements to system during the phase of the customers testing and prototyping. They need to maintain strong attention to detail for spotting errors in data or coding having a knowledge of microstrategy and data architecture is a set bonus and a little efficiency in written and oral communication.

Some skills never harmed anybody now the skill set that I have mentioned is enough to make yourself fit for the position of a tableau developer. There are many opportunities available in the it market for candidates who acquire the skills that I have just mentioned but the road to acquiring these skills is a long one and we added Eureka want to help you out with this here. You can learn at your own pace and take your time to make yourself industry ready in tableau. this program provides structure and guidance and is curated specially by industry experts which covers extensive Concepts such as data blending creation of charts and level of detail expressions using versions of Tableau such as Tableau desktop Tableau public and Tableau reader this also covers integration of Tableau with our and Big Data having said that let's discuss a little bit about the future of Tableau now the reason why tableaus, Hands out amongst all the bi tools is that there is a wide range of connectivity options Tableau can connect to any data that you can possibly think about starting from spreadsheets databases and even big data you can access warehouses Cloud applications like Salesforce and even connect to Cloud databases like Amazon redshift.

It has a web data connector and can be used to pull API directly from web in order to connect with desired data source apart from that in this world. Of prevailing Big Data many organizations that store wangle and analyze data choose Hadoop as their platform of choice Tableau authorizers businesses to easily and quickly identify valuable data in their expansive Hadoop data sets and removes the need for its users to have the knowledge of query languages that makes engaging with big data more feasible for stakeholders, but now natural language processing and machine learning enabled data are two things that tableau. Is focusing on it is molding itself with new technologies to enable futuristic approaches to view data. It is going to launch a hybrid data connectivity for cloud and with Tableau.

Another Advantage would be a new life query agent that will act as a tunnel to on-premises data, which will obviously expand the caliber of Tableau. So go ahead and get started with tableau. So before we start to go through the different questions, you can encounter during table related interviews. Let me just sort of like, you know, tell you the background whether you have made the right choice by choosing to learn Tableau or not. So here in this particular slide, you can see the future of Tableau as a software this particular chart, which you can see is known as Gartner's magic quadrant.

This was published on February 2015. They have published a recent one. in February, 2016 and Like previous magic quadrant stab you again is the leader in terms of ability to execute its again in the leaders quadrant. It has come down a little bit in terms of the Visionary aspects. I don't know why maybe you know, I saw Microsoft was right here if you see the 2016 slide but still table is very much the top contender in terms of ability to execute its at the highest position. So the future of travel definitely is very bright for the fourth consecutive. It has been way ahead of its competitors in terms of ability to execute and in terms of completeness of vision. Again, Tableau is a strong Contender. It's lying right here in the leaders category. These are challenges. These are Niche plus these are Visionaries challenges are those who have good ability to execute but, you know the different variety of work which you can do through these software's are limited.

So they have specific Focus Visionaries are those Those which have long-term good prospects, but it's difficult for you to execute your work in those players who are lying in this Visionary box leaders on the other hand. They can perform a variety of tasks for you and you can do those Works relatively easier and Tableau is lying right here at the top of the stack in terms of ability to execute job Trends in terms of job Trends. Tableau has shown very good progress. So here you can see the Of jobs, which are related to table you have increased exponentially. So, this is January 2015. I think this might be January 2016. You can see the demand for tablet professionals have grown considerably and the national salary trend for Tableau again shows that that the salary trend for Tableau is increasing. So it's worth our time and effort to learn Tableau and work in this particular field. One thing which I can't Tell you from my perspective being a tableau developer myself. I can tell you the job satisfaction level is going to be high because it's fun to work in Tableau.

So you get constant feedback. So from job satisfaction perspective, I can tell you you will get instant feedback while working with Tableau. So everything is visual you all the components. You invoke all the components you work with the output which you get everything is right in front of you. So, you know, it's pretty exciting to work on Tableau as far as I'm concerned I can vouch for that and some major companies which are using Tableau include Cisco Google Yahoo. LinkedIn Facebook YouTube as you can see the list here. Okay. So I was telling you about major companies which are using Tableau and you can see you know, many big players are using Tableau. I've worked across multiple multinational companies many of them were Fortune 500 and many of them were using Tableau. So I work in John Deere. I worked in Rio Tinto all of them a Fortune 500 companies all of them. Them are using Tableau.

So industry adoption is very high. And these are two top Contender when it comes to visual analytics. We have Tableau and we have click View and here is a comparison between them the strong suit of Click view versus Tableau. So both can analyze Big Data Tableau as you can see I can I unless billions of data analyzed millions of data ETL tools are available in qlikview. ETL tools are not available in Telugu a so-so. That's one constraint which you will face into a blue and you have to do your data preparation most of it outside of taboos environment.

So if you are well aware of any database if you are well aware of SQL coding if you are well aware of Excel or access if you can manipulate your data well through some other tool then using Tableau. You can visualize that data click view on the other hand offers you some ETL tools so you can manipulate your data within click View. As well click view versus Tableau. Another difference is qlikview is very technical.

Okay, it's much more complex as compared to Tableau Tableau on the other hand is very intuitive very user-friendly and to work with Tableau to pick up Tableau to learn Tableau is much easier as compared to click View and once you have you know, so while working with qlikview dashboards, you have to keep your and objective in mind you have to do all those preparation going back and changing your We take extra effort Tableau on the other hand provides you the capability of rapid-fire analysis. Okay, so you can just push in your data within chair blue. You can slice and dice the data you can pivot this data. You can experiment with different kind of visualization.

So you can work on the Fly very quickly at any point of time. If you want to go back and change some component of your visualization. You can do that very easily that is a challenge with click view. Okay. So if you have to build that context you have to do thorough Planning these kind of constraints are there with click view if you are looking for a very easy to work software. Tableau is the tool for you. Now, let's get into the Crux of matter. We are going to pick up certain interview questions, which we have framed for you. Some of these questions has been asked and raised by some of you and we have tried to answer them here. So the first question why is there a need to go for Tableau in spite of having huge number of Open Source and less costly visualization tools? The some of the reasons which I have mentioned this this particular question has been asked by one of you. Okay, and this question actually makes sense. Tableau has a lot of competitors. Okay and just guys just give me one quick second into for that matter.

So the answer is first of all Tableau is very very easy to use dashboards are simple to build even for a newest somebody who has never really worked with the visualization tool can learn Tableau easily. That is one big reason. Also, even for an end user from an end user perspective Tableau is very easy to use so they have all the filtering capabilities parameters all at their disposal.

You can provide your end users with these features and they can play around with the dashboard you have created for them. So adoption from a developer's perspective and from an end users perspective. Both is very high rapid visualization drag-and-drop feature feed. So Tableau has this drag and drop feature which you can use for creating visualizations. Rapidly, even very very complex dashboard to not take months to create which they usually do in some if you're going to try some other visualization tool within Tableau even complex dashboard.

You can create in matter of few days. Okay, not even a week. I'll say so it's just drag and drop and you have to be aware of some Advanced feature from time to time you have to use them. But mostly it's just the basic feature which you are going to require most of the times now the charts which Tableau provides you are very visually appealing Okay. Tableau is based on best practices of visualization. It doesn't provide you 3D charts 3D charts are not good. Okay. So this has been proven through different research that 3D charts and other those fancy charts. They are not good for visualization simple and Visually appealing charts are best means to present your data and Tableau is based on best visualization practices. It allows you to create visually appealing charts with vibrant automatically generated colors. That is another reason Tableau is very intuitive. You can customize you can modify your visualization to a great extent and you know, once you practice Tableau enough and I'm not talking about, you know months and months of practice. I'm just talking about chocolate several weeks of practice.

If you work like, you know, do three four projects in Tableau, it will become very very intuitive to you because all the components which you utilize to modify a chart are right in front of you so we have those shelves and cards when you need to drag and drop your Your values if you want to show Legends, they are right in front of you, you know how to do that. And if you want to change the color scheme the option of changing the color scheme is right in front of you so table is very intuitive. I have tried to use click View and it was not that intuitive at all. In fact, if I compare it with Tableau, it doesn't stand anywhere in terms of ease of use and intuitive – also Tableau uses Excel like formulas and it's easy transition for many because many people work on. Miss Excel and those of you who have already worked on Ms.

Excel and since you know, I'm assuming all of you have learnt a blue so it will be an easy transition for you many of these formulas follow the same syntax within Tableau and Excel for example, if else formula the syntax is more or less similar. We have date the formula all of them are using Excel as well as Tableau. So easy transition, then we have a active user Community. Now, this is one feature of Tableau, which is not very commonly mentioned, but I am going to mention it here. Okay. So as a developer, you know, if you have worked on some other platforms some other programming language, you may realize that from time to time we have to depend on other users user Community. We have to Google our answer. Okay, we cannot just remember all the keywords all the technicalities of the software ourself either. We have to keep like a thick reference book or reference manual or we Google our answers up.

So there is a need that your software should be supported by an Was a community and Tableau serves that feature, you know W has that feature. So there's a very active user Community which supports table. Umm. Okay, then Tableau has in memory of bi platform enabling High scalable and Rapid visualization. Tableau is pretty fast. If you are working with large chunk of data table is not going to disappoint you I mean, of course if you're working with billions of rows of data, you have to optimize the performance of your dashboard. The response will be relatively slower then when you are working with smaller did Yeah, but just compare Excel and Tableau.

There is no comparison Tableau just doesn't compact you data. It actually handles the data much more faster. Okay ability to apply filters and date range on the Fly. Of course, you can apply filters. You can filter out date range. You can show those filters right on top of your dashboard and your end users can play around with those filters as well. Not only that you can click on one portion of your chart and filter down other charts as well.

If you have made a dashboard there six seven different charts, you can click on one particular chart or the charts will filter down automatically if you have enabled those features and Tableau works on user feedback. Now again, this particular benefit of using Tableau is not mentioned commonly. But what I have seen is have been using Tableau for quite some time now and I started using it way back like I think from version 5 or 6. I do not remember exactly, but it was lacking a lot of features back then and users were constantly providing. Feedback that this particular feature should be available that feature should be available. In fact, we got this survey from Chapel where we provided our feedback and several of those feedback which we provided.

We saw that in action after a few versions. So now Tableau has the integrating feature with our Tableau provides floating filters, you know, floating containers earlier. These features were not available but Tableau works on user feedback and incoming few years. We may see most of the desired feature are We'll interview right now. There are some limitations as I mentioned like ETL tool is not there. Who knows we may see some ETL tool with interview itself in couple of years from now. In fact Tableau has started providing some data manipulation feature in the latest versions. So for instance here, these are different marks and these are different cards and shelves which you are using to control your chart to further modify and customize your chart. Okay. They're right here in front of your eyes in. Excel you probably have to right click and go to certain particular, you know areas of the pop-up window, but here these components are right in front of you and you can just click here.

You can modify the text. You can set the alignment all these things you can do right in front of yourself. So it's pretty intuitive. All right, let's move on to the second question, which we have how to optimize the reports for better performance here. I have put down certain pointers. This is not a limited list. This is not a complete list as such. Okay, as you keep working with Tableau, you will realize that there are many more options which you can employ to fastin, you know, improve the performance of your dashboards. So a recent test which was run side-by-side on two different machines running the same data pull from Tableau server Internet Explorer 7, return results in 11 seconds Firefox return results in 3 seconds. So the browser which you are using that makes a difference, okay? So when you are working with Tableau server, it's preferable to tell your users to use the latest browser, which is the fastest one. Okay, so it does make a difference second tip which I would like to share with you if complex calculations are needed in Tableau consider creating a dbms view that does the calculation the database server is usually more powerful than the desktop.

Okay. So if you are going to filter your data, okay, if you're going to create some aggregation you can do that. Within your dbms as much as possible. Are you aware of granularity of data? Okay. So keep in mind the granularity of your data. If you feel that you are not going to use data at a very granular level at a very small level. If you are always going to use it at an aggregate level. Go ahead aggregate your data within dbms itself and then pull that with interval. Oh, it's always going to help. Okay, so tell you doesn't have to do all that hard work of aggregating your data.

Let's see if I can give some examples. Example, let's say you have daily transaction it or maybe per transaction data. We are collecting data for a superstar. Let's say we have Walmart and we are collecting data for Walmart. We have data for each transaction, but the analysis which we are doing that is happening only at a product level or maybe at a daily level or at a monthly level. Okay, you are never going to go to particular transaction level and analyze the data, you know that already. Okay. So what you should do is you should I would summarize your data at dbms level and then pull that data Within Chapter. It's going to help does that make sense and consider the use of pre-computer aggregated summary tables when large data set are required. Typically when summary summarize when used okay again, same point, you have to sort of like keep in mind the granularity of your data.

If you are working on calculated fields, which are not going to change frequently, maybe row-level calculation some if-else calculation which are doing okay. You can do that right within the BMS if you are aggregating your data summarizing your data do that outside of Tableau. It's going to increase your speed. Of course, these points are applicable only when you're working with large data sets may be Beyond, you know, 10 million rows or something then it's going to improve your performance drastically. Otherwise if the number of rows you are working with is less than 1 million then you know tab you will be able to handle that seamlessly.

It's not going to be a problem at all. Also, you should turn off automatic updates. Whenever possible so, you know Tableau doesn't automatically updates your data, you know, instead of correcting making a live connection make an extract updated manually whenever possible. If you do not want Auto refresh or the data that's going to speed in of the things. So next Point use as few data sources as needed to achieve your analysis and remove any unused data source, we have this tendency, you know of connecting to different data sources as an analyst I can say this I do it from time to time. I've seen other people doing it, even if they do not need a particular data source still they make a connection to it. Okay, because they might need it in future if you're working with big data do not do that.

Okay connect to only those data sources which you are going through use which you are sure you're going to use and use extracts. So when you extract the data, you will actually get additional features like, you know, you can use certain formulas. Tableau is going to optimize your data to improve its performance. So use of extract is going to increase the performance of your dashboard when filtering try to avoid the exclude option. So when you're filtering the data, there is this exclude option. If you use that Tableau will perform a bit slow. So went exclude option is used Tableau will scan all the selected data and then it will exclude it selectively. Okay, so try to avoid the exclude option rather use the include option. And use Boolean calculation whenever possible with Alias. Okay. So how to do that how to use Boolean calculation whenever possible. Okay, so I'll just repeat what I said. So couple of tips which I was sharing you can create an extract of your data that's going to speed in up your calculations because Tableau optimizes its extract to make the dashboard perform better.

So always create an extract whenever possible you can refresh an extract. So refreshing your data if you're worried about refreshing your data, whether it's going to work or not. That extracting it it's going to work. You can refresh the extract as well and when filtering try to avoid the exclude option because when you use the exclude option Tableau will scan all the selected data and it's going to slow down the calculations a bit. Okay slow down the process a bit rather try to use the include options. All these points are applicable when you are working with big data set and use Boolean calculations whenever possible with Alias. So for example, if you are using an if else formula, okay, if you are using an if else formula, you're trying to let's say I analyze if let's say if the age if age is less than 25 years or if age is less than 18 say not a result else say adult. Okay, if that is the formulae trying to build just build a simple formula age less than 18 you will get results in form of true and false.

Okay, and then use Alias to rename true as not adult and Falls as adult. Does that make sense? Let me just quickly show that to you through an example how that works. Sure. Let's say we have fixed salary of different employees. Okay, so we have different employees and their fixed salaries. So some employees are getting above 10 lakhs. Some employees are getting less than 10 lakhs. All right, let's proceed further. There are other ways in which you can optimize your formula use else if rather than nested else if in your Logical statement here is a Formula you can see if region is east and customer segment is equal to Consumer then call it East consumer else.

Now. This is the start of another if statement hence to and statement here. Are you able to recognize this? So one if statement another if statement this is nested if one statement within another okay this one Be slow. You can write it like this instead of giving this space between else and F and making it nested. Just give the next option as else F. Okay, check the second criteria here. We have just one if statement there for just one in statement. This will be much faster. So these are some ways in which you can optimize your if-else statement. Let's say you do not need the details of time stamp level.

Okay. So in that case use today do not use now for smaller data set more Less you can use today or now depending on your preference. I mean, it doesn't really matter but when it comes to large data set, if you're performing a lot of time related calculation based on today's date, you do not need time stamp level details don't use now use today. Okay, so that will restrict the scope of calculation. It will be faster. Also, you should logically optimize your calculation. So for example here take use of, you know, make use of logical calculation these two statements these two if statement are going to produce the Exactly same result but this one the one at the bottom will be faster. Okay. So what does this statement do this statement is checking if the sales is less than 10 then it is categorizing this categorizing the sale as bad else if the sales is between 10 and 30, then it is saying okay if sales is greater than 30 then it's saying great.

No more many people will write the statement like this and there's nothing wrong, but if you want to make it perform faster, Then you can just ignore this statement. This is the default statement. Okay, you do not need to check for this condition check for first condition if sales is less than 10, then it's a bad sale if sales is greater than equal to 30 then it's a great sale. Otherwise, it's like somewhere in between we have already on this is a logical statement, right? So we make use of logic and we just skip this entire check and that will make our code run faster.

Next step when using extracts and custom aggregation divide the calculation into multiple part place the roll level calculation on one calculated field and the aggregated calculation on a second calculated field and then extract can optimize the precomputed row-level calculation. So for example, if you are trying to calculate the sum of salary of all male employees or average salary of male employees versus average salary of female employees you can The formula something like this.

You can type in the formula something like this. You can say sum of salary whatever, you know, if if gender is equal to mail and then salary else 0 so what are you doing here if gender is male then you are taking the salary of the person else you're taking zero, you're not taking the salary of the person then you are summing it up. So in one way you are performing a row level calculation and then you are aggregating it. Okay, this tip this particular tip tells you if you are doing something like this break it apart into two different pieces create two variables type it like this this entire thing will become a new variable. Let's say we'll call it male salary then you will say equal to this will be the formula of mail salary. And then you will just perform a sum of male salary. So they'll be two variables if you create two variables and then take an extract table is going to perform the row level calculation and optimize it and then aggregations will be faster remove any unneeded dimensional measure from your palate next point.

So we are working on some particular visualization. Let's see this one and what you will see from time to time. They'll be something lying in the detail section, which you are not using. Maybe you were experimenting with your view you try to create three four different kind of visualization, and then you landed upon the visualization of your choice. Okay, but what happened in turn was there were some there were some information which got left behind in the detail section. It has happened with me, you know at from times to times I go back to some of my visualization and I see there is something lacking in the detail section.

I never intended to put it there, but then I was experimenting with my visualization and bye. Steak. I left in a particular unwanted filled in the details section. It didn't hamper my visualization. It has no effect on my visualization as such okay, but it's still just lying there sitting there making my dashboard slower. So in those cases, it's actually good to revisit your shelf and see if there is some extra item, which is not needed typically in the detail shelf. You will find some of those things just remove them. Okay, so there are a lot of tips which I shared with you. In terms of making your reports perform better performance optimization therefore the tips which I can give you for example, if you are making a tabular structure interview, you know, you're not really creating a visualization rather creating a tabular structure like this. This is table and people are supposed to scroll through this table and you know look through the data so we have this tabular structure.

If you are picking any of these three, make sure The scrolling feature is not taking you very far down because if your table is very very big. It's going to give a you know, terrible hit to your performance. Okay, so just make sure if you're doing something like this create a hierarchy enable plus and minus features, especially, you know, people create huge table when there is hierarchy in world. So for example, Global sales broken up by Regional seeds broken up by country level sales, bro. Open up by state level sales broken up by City level state sales. This will turn out to be a huge table and people have to really scroll down create a hierarchy let people click on plus and minus button, you know expand or collapse The View because that we use going to slow down new dashboard considerably.

So that's another tip. I wanted to share. And then we have some context filters. Any one of you is not aware of context filters their context filters. They are also used for optimizing the performance goes from X to X. So there are many different features. You can use to optimize the performance of your dashboard. Okay, third question canterville, you create operational report where data changes every second and also it looks like the drill down the table data that Tableau shows as picture format and cannot be refreshed real.

Time I was not able to comprehend this question completely but I am assuming this question actually means that how you can actually refresh the data table. Okay. So there is a scheduling n sort of like scheduling tasks which you can perform in Tableau that is done on server side actually. So here is a link which I provided you will get this PPT. And once you click on this link, it will give you step-by-step guidance on how to schedule reports. There are some predefined schedules. So for example, Can enable your report to refresh every 15 minutes or end of a particular month or every four hours. So there are different predefined timings in which you can schedule the refresh of your report. Let's move on next question how a real-time Tableau project will be in an organization from development to publishing.

Okay. So it actually differs from condition to condition, you know from situation to situation. There can be different kind of tableau. Projects you can encounter and I'm just purely answering this based on my experience. They can be other situation as well. Okay, Tableau project can either be a migration project a migration project means your report was existing in some other tool. Maybe it was an Excel report. Maybe it was a business objects report and now you have been given the task to migrated to table this this real stuff, which I'm talking about, you know, and you will probably encounter situation out of these four only there might be Other situation which I have not encountered and you may okay.

So these migration projects they are easy to work with and they are profitable. So by profitable what I mean to say is people can people will actually get to compare your report versus the older port and they will be able to see the difference and you will get a lot of praises and I've got that my colleagues have got that in past. So these are like, you know good projects to work on Plus. Amounts will be crystal clear because you are migrating something. All the requirements have been noted down. There's a practical application sitting in front of you which you have to replicate in Tableau and make it better, of course visually appealing and you know jabu's fast engine slicing and dicing ability all these things you have to apply these projects will be easier to handle and mostly it will when you lot of accolades and appreciation. Then you have brand new projects brand new projects as in like, you know, business doesn't know what it wants you have.

Data sitting in front of you maybe business has asked you to create a particular report, which was not existing earlier and you are doing it in Tableau. This is also a good thing to work on and you will in turn learn a lot of things what hear what you have to do is really experiment a lot with visualization just be creative think about you know, what all visualizations you can show your dashboard. So again, these kind of projects are there then we have ad hoc reporting projects where you are given exact set of requirements small.

Projects may be okay and you have to create a small Tableau dashboard in order to fulfill those requirements. It may be required for one-time use. Maybe it's going to get used for only one week and then scrapped. Okay, so these ad hoc reporting products are also very common across different Enterprises. These are done in tell you why because table is very fast and you can actually create reports like you do in Excel pivot tables and you can actually create reports that quickly interval. So a doc reporting is again a very common tasks in Then exploratory research to guide predictive modeling if you're working on full-scale predictive modeling projects. So Tableau is going to be a very handy tool for you. I'm from predictive model modeling background and I can tell you that within predictive modeling there can be lot of permutation combination.

It would really help if you know your data best. Okay interview helps you do that Tableau gives you output in a very practical and I mean the entire feel of data you will get while working with them. Blue, okay, so you will know in and out of your data. Once you come out of the Tableau environment and slice and dice the data, you can go to any level in Tableau. So exploratory research to guide predictive modeling is done quite commonly interview. So you'll pull your data you'll create scatter plot. You'll create some bar charts and you'll see how the trend is moving and then you will deploy predictive models where you find interesting Trends or where you see there is potential of predictive modeling to bring out some good results. Okay. Otherwise it's difficult to No, just go ahead and deploy predictive modeling and then come out with no results and try some alternate approach. It's a slow process and very intensive courses. So w is going to help you there. These are different kind of projects. You may encounter and based on the kind of project you are doing the approach which you're going to take will vary so I will talk about that briefly in my next question not this one.

I will come back to this before that. Let me try to answer this particular question. Now this has some relation to the question which I just discussed the different projects which happens in Enterprise and what are considered to be the components or or the approach which you should take to create the best possible dashboard. So first thing involve business people in dashboard Design This is a softer aspect. This is not anything technical but it hugely affects the success rate of your end product. Okay specially if you are working on These kind of projects if you're working on migration project or brand new projects.

It's imperative to involve business people have recurring meetings with them. Okay, probably if your project is going to be like two or three months long have recurring weekly meeting may be okay or maybe twice a week show them your progress take feedback from them and then keep on working. They will provide you the necessary business context as well. So my grave Great Migration project and brand new project in those cases. You should involve business people and seek their feedback and use an iterative dashboard design approach which means build your dashboard certain components of it show it to the business gather their feedback and then work on it again.

Okay, this scope of success the chances of success will be much higher then allow drill-down capabilities within dashboard. You should always allow drill-down capabilities within dashboard rather than creating separate. For each different condition rather than creating distribution of mail. Let's say we are working on a HR dashboard and we want to see how many male employees are working across different departments. How many female employees are working across different departments? And then how many male employees less than 25 years of working in different department? How many male employees between 25 to 35 were working in different departments? So we have all these different permutation and combination one may feel tempted. Added to create, you know, multiple different histograms. Maybe for each of these situation do not do that. Use the drill down capabilities of dashboard create a pie chart for male-female create a histogram for age distribution and then create a bar chart for number of employees working in each department show your uses that they can click on one section of the pie then another section of the histogram and then the bar.

That will Auto filter. Okay, so they can choose male employees between is of 45 to 60. They can choose female employees between age of 25 to 35 and they can see the numbers how many employees are working in each department these drill down capabilities of Tableau. You should utilize to the fullest. Also, please include actionable informations some information which might be very relevant and interesting to you might not be actionable to your end users. Okay, you may assure them that your sales is decreasing. They might be knowing it already or even if they're not knowing it. It's just going to panic them. What is actionable information. Why is the sale decreasing? Okay, what can probably be done to improve those sales where our Still going. Okay. What are they doing? Good. Okay, so these kind of action and actionable information if you show then, you know the adoption or the success of your dashboard will be high don't include too much. So don't overburden your dashboard. Keep it simple. Keep it non-cluttered. If you want create multiple dashboards in your file. Okay. Don't put too much of information within one single screen show relevant filters relevant filters means in Intel.

Ooh, you have this feature of creating relevant relevant filters. I hope you are aware of it. Right? So for example, I can use floor number as a filter here float number and these flow numbers are appearing. This is showing me different employees and which flow they are working on if I want to see only those employees who are working on 4th floor. I can select this filter and then we have this Department's name. Okay, so we have all these different departments now thing is some of the department they are sitting on particular for floor. So for example, CEO office see you office is sitting on floor flow development development is also sitting on 4th floor. HR HR is not sitting on fourth floor, but still it is showing up in this filtering keep filtering option show relevant filters. Okay. So what you should do is click on this drop-down and Choose this thing only relevant values. This is a big confusion point for and users.

They will try to choose a combination which doesn't exist and then they will get scared that there is some problem with the data. They will come back question you they'll get frustrated. So these kind of things happen. This is this very simple option, which you should always keep in mind show only relevant values. Now if I'm going to choose fifth floor, I will see different departments if I'm going to choose sixth floor. I'm going to set different department. Um, If I'm going to choose all see all the Departments so show relevant filters. Okay. This is a good practice keep an eye open for color blindness. I was very ignorant towards this particular fact till I learned it the hard way. I created this dashboard. It got sent across to multiple different leaders and the top leader. He was colorblind. I didn't knew I was not careful enough and you know that dashboard came back to me. I Add to make it colorblind proof. So if your dashboard is being shared across multiple different users may be like you're 50 or a hundred different user assume that some of them probably will be facing this difficulty.

Okay, and just adopt some technique through which you can count the color and brightness. What are those techniques you can use Shades instead of vibrant colors? Okay. So if you are creating some, you know, if you're creating a bar chart like this maybe Like this. Okay. Try to give [ __ ] here. You can see these are shades of the same color two different shades of almost the same color. This can be interpreted by a colorblind person. Okay, if you if you give too dark colors of different shades may be green or yellow color blind person may find it difficult to comprehend.

So just keep an eye open for color blindness as well choose between percentage and real numeric value. So from time to time you have to show values as percent. Age from time to time you have to show real numeric values if we have three different departments. Okay, one department is testing Department, which have 500 different people working. Another department is HR department, which have 50 people then we have pmo project management office, which have let's say 10 people only. I want to see. Bifurcation of male and female in these three Department. I want to compare which has a healthy gender ratio. What should I show real values or percentage values percentage values is what we should be showing right because you cannot compare apples and oranges. Okay. So if you say there are a hundred female in testing and they are only 40 female in HR that will present a very wrong picture because hundred female out of 500 in testing that's like 20 percent and 40 female out of 15 HR.

That's like 80 percent. So there's no comparison. The gender ratio is probably you know, the female ratio is much more higher in HR. So sure you need to show percentage values. So just think carefully about it the example, which I gave to you was pretty apparent in real world. It might not be that happen. So you have to Think of it carefully whether I should show percentage value or numeric value and follow consistent and differential color coding is for example, if you are building HR dashboard, you are showing male female in a pie chart and then male-female again in some hundred percent stacked bar. Try to follow consistent coding for them. It should be consistent and also it should be differential. So for example if you are using Yellow for female then don't use yellow for managerial employees. Okay. So here is an example here. This is managerial employees. These are female employees this code for female.

This is code for people managers. Okay, those who have some team reporting to them whenever I'm going to use people manager in any other dashboard as well. I'm going to use this color only whenever I'm going to use female anywhere else maybe if I'm going to say, let's talk about terminations. You know, how many female got terminated how many male got terminated like that? So I will try to use the same color scheme. So be consistent and differential show applied filters for user. He's okay. Now what happens is if your users have too many different filters and specially if they are working across multiple dashboards, if you have provided them six seven different dashboards and somehow you have linked these dashboards so there Line Filter in one area and there, you know actually looking at the dashboard in some other place.

It's a good idea to actually show them what filters they have already applied. How can you do that? Let me see if I have an example here. I'm going to open up an example which will show you. Okay, nevermind. I'll create the feature here itself. So let's say I'm applying a filter on those employees who are working on 4th floor and can then those who are working in testing and those who are working as senior associate. Okay, if I want, you know, it's actually a good idea to show your end user which filter they have already applied. How can you do that? Maybe let's do it your I'm going to expand it a little bit.

I'm just going to edit the title. And you can sin. Okay. I'm sorry. Not you. Let me just with that some other place. Let me do it here so you can say gender gender is equal to gender and then you can say let's say Department name is equal to D PT is equal to department. So like this you can create sort of like labels here in a better way. Of course. I'm doing it very quickly and then if people are applying filter lets somebody apply the filter on pmo. They'll be able to see right here that the pick the filter they have applied. They have chosen Department to be pmo. And then let's say they have chosen female. So the current filters the screen which they are looking at is for female gender working in pmo.

It's a good idea to actually enable it. Yeah, and especially if you are using action filters, then you can actually apply filters across different dashboards. Okay, so I can choose female here and it's actually going to filter out some of my dashboard, you know, some of my charts in this particular dashboard. You can enable those filter features also using action filters in that case. It's absolutely imperative to show these filters. Okay what filters they have applied otherwise people will get confused. They'll absorb some other information. They'll make different. Predation or affect the might not even be aware. They are looking at only female population. Okay, they will assume that it's for all employees.

Okay. So show applied filters for user. He's it's a good idea and test your dashboard on different screens. I would recommend using this particular feature of Tableau. Go to dashboard keep a fixed size. Okay, do not never ever ever ever choose this option. This is Is suicidal. Yeah, and that's the word which comes to my mind automatic. Never choose this if I'm going to project this through a projector on a you know, a screen it's going to look different if I'm going to transfer it onto some of you you have a different screen size. It's going to look different in your case. Some of the charts May blow up some charts will become exceedingly small some charts will become exceedingly big this option is never a good idea always go for exactly or choose like a no Defined exact size and also it's good to check it on different screen specially if you are going to transfer it to different users now many of these tips are applicable for these two projects.

So for migration projects and brand new projects, especially if you have more than no.3 for users who are going to use that particular dashboard, make sure you are following those tips, which I shared with you maybe for ad hoc reporting which is going only to one particular. Ziggler user or for exploratory research which you're going to do for yourself. You might not require those steps. All right how to do effective data blending. This is another question which we got so for data blending, of course, I mean, it's a more complex Topic in Tableau, but for Effective data blending, these are some points which you need to keep in mind when you have data blending done. Your primary data source will show up in blue and secondary data source will show up in Orange here. You can see this is secondary.

This is primary. What if I want my dashboard to be viewable on mobile devices. So in case if you want your dashboard to be viewable on mobile devices there might be some option here. I have Let's see we have iPad. We have iPad landscape. You can also specify exact size. So here you can Define exact width and height and you can optimize the view for mobile devices. You can do that and just in case if I make it large enough, let's II make it like, you know if I increase the height to 900 I'll still At the scroll bar. Okay, so I can scroll through my dashboard. Only thing is it's not going to fit on one single screen, but the aspect ratio of charts is not going to change that is one huge Advantage which I will be getting if I'm going to choose automatic.

My aspect ratio will change this will become broader. Some will become smaller. You know, that's going to become very challenging. How to connect to a database okay, if you want to connect to a database when you open up Tableau or let's say you have created your dashboard and you want to connect to a database here you go to new data source.

If you have access to Tableau Data server, you can go to the section. So some Enterprises what they do, they'll build their own Tableau Data server, and they'll provide you access to that data server and you can connect from here else. You can choose new data source. I am using The public right now, so I do not have option of connecting to a database but you will see those options available available here. Okay, and once you click on the option, let's say you want to connect your Ms. SQL database through Tableau. Okay. So let me show you how it looks like. So if you want to connect from tablet Ms. SQL, first screen, which you will see will be something like this. It's kind of hazy. I'm not sure if you are able to see it. Well, so you will get all these different options when you are working with Tableau desktop, you'll get all these different options.

You can choose anyone as per your preference. If you want to connect to Oracle database, let's say you want to connect to Ms. SQL database. Okay, then you will see a screen. Like this. Yeah use Windows authentication or username password you can provide and then you will see a screen where you can actually write in the SQL code. If you want to you will see all the tables which are available. And you can also write in a custom SQL if you want to something like this. So let's say you have a sequel Builder. It's a you have MS you're working on it. He's our you're working on Ms. SQL database and there's a sequel Builder which is provided with those software use the SQL Builder create your SQL code copy the code Connect using Tableau to that database paste your code here test your connection and then boom you're done.

Then you will open up table. You'll see feels like this. They'll be coming directly from your database. And of course you will have that option of connecting life or creating an extract like we do for Excel right you can either make a live connection or you can create an extract. So these options will be available. It's fairly simple. It's not going to be complicated. They'll be one screen after another choose your database provide your credentials choose whether you want to live or extract connection provide your SQL query now, we are blending data.

First thing you need to know and of course, I'm sure most of you might be knowing it already. There's a primary data source. There is a secondary data source primary data source looks up in blue. Secondary data source shows up in Orange the first field which you are going to pull up into your Tableau environment in your particular sheet that is going to Define your primary data source. There's no set primary data source. Let's say we have sales and promotion sales have 10 Fields promotions have 20 feet. I click on this container container for sales. Okay. Let me just see if there is a dashboard.

I have pre-built which has data blending feature. So this is data joining by the way, you can see here. This is data joining. Let me just connect to another data source new data source Excel. and I'm going to connect to a Char B1 and I'm going to connect to let's say performance rating. Okay performance rating and I will go back to the sheet.

So I've connected to performance rating. Now I have two containers. Okay one and two if I'm going to build a view let's say I take last rating into this field. This becomes my primary data source, this becomes secondary. Okay. Now if I'm going to pull it's a fixed salary doesn't make sense. I don't know if it's going to work or not. See these blue and orange symbols. I can do it twice over so I can do it the other way around I can pull something from HR first this one and then something from here.

Okay, and it's going to go Almost similar. This is primary and the secondary so primary and secondary is actually defined by the order in which you are pulling the data. There is no fixed primary. There is no fixed. Secondary a default blend is equivalent to a left outer join. So, you know all of you know, what is left outer join. We have two tables one on the left one on the right. It's like, you know, imagine when diagram set theory. Okay, all the components all the data from left side of being pulled and oh, Only matching data from right side is pull. Let me just show you left outer join.

Let's see if you hear this. So left outer join everything from the first table is pulled and only the matching record from the second table is put this is default blending. If you you know here what I have done. I have pulled something from HR V 1 and then I pull something from performance rating. This is left outer join anything which is in HRV one will be pulled and only the matching records from performance rating will be put this is a right outer join everything from performance. this bolt and only matching record from HRV one will put okay so you can actually sort of like simulate sort of, you know left and right joints by switching what your primary and secondary data source if you filter out nulls, okay, you can simulate an inner join as well but no full Outta joint that's not available through but blinding how to create parameters if we are using two or more data sources one data source have different data and other have different data how How can we use a global filter from two different data source, if you are using blending, okay filtering if you apply filters, it's going to work.

Okay, but you can also use the parameter feature. So what you need to do is probably create a calculated field put that in the filter section and it's going to work. I'm not able to pick up the proper example here. Let's see. We have employee ID. And we have here we have let's remove fixed salary. Let's say we have different departments. Okay, you have to enable the blending feature on the correct field. And in this was just part of the tip which I shared with you. We have to still come up to that level. Okay, so while doing and I'm just going to answer your question on filtering. Let me just complete this and then I will show you the example on filtering. So sometimes you see null value and this star symbol. We just make sure that you are clicking on the right chain. Okay. So when you are doing data blending you see this chain like symbol and base.

On the field names so we have last rating in our primary field and in our secondary field both and that's where Tableau is suggesting. You can create a blending based on last reading or you can create a blending based on employee ID. Okay, the correct field should be employee ID and you should enable that. Okay, so just make keep an eye open for that. Otherwise, what will happen is you will be able to see this star symbol.

Okay. It means that your data source do not connect a contain enough information to blend. Okay, so just be keep an open eye for that now I have enabled. A connection between employee ID between performance rating in HR now, let me just pull up number of employees. Okay. I'm going to pull up number of employees and I am going to apply a filter on Department. Let me apply a filter on Department. Okay, so you will not be able to see quick filters. You can just drag it here and then you can choose certain departments. Let's say I choose development and HR only say apply, okay. Okay, and you can actually apply this for 10 then once you have dropped your feel in this filter section, then you can right-click and choose show quick filter.

And then it's going to work just well. This is one way another way is you can create a parameter across and the parameter and a calculated field in in your particular know if both of your data source contains the same values you can make use of parameters as well create a calculated field use them in filter. Like I have done here Department. Though it was not a calculated field. There was no parameter involved but you can apply filters in Blended data. Okay, last rating and department is last rating if you want to show it as quick filter you want to show only high performance may be in development. It's going to work. Well, no problem. If you want to go into this particular thing and you know, you want to choose some flow number and show it as quick filter.

It's not going to be straight forward you have to Drag it first and then choose the values which you want to show say, okay, and then you can show it as quick filter. And now I can apply certain filters here and let's say I choose all the Departments so you can see those departments which are working out a fifth floor. They are showing up. It's perfectly fine. You can do that. You have to be a bit careful. It blending is a bit complex. I do agree and sort of like, you know irritating also from time to time, but It works. Okay, and you have to keep in mind that changing the blend of the fields, you know, the order of blending or Blended Fields will change the scope of your analysis.

Okay, this is something which you need to keep in mind. So let's say we have this data set. Let me just open up a data set for you. This is going to be a big one. So I'm not going to open up the main table, but I'm going to open up the lookup table. sure, so we have Data for different airlines. Okay, which are flying in us and then we have a lookup table for airport. So an airline will have a departure airport and Airline will have arrival Airport. In the main data, I'm not going to open up the data because it's a big data and it's going to take some time. You will see in the main data. You will see just these codes iata codes, but in your dashboard you would like to show the airport named probably okay, and also which state the belongs to which country they belong to and these kind of information you would like to fetch in your dashboard. So there will be an arrival airport and they'll be a departure airport if you blend your data.

Between departure and this thing, you know, I ate here you will get description of departure airport. And if you blend your data between arrival and this particular key field, this will actually show up your arrival airport. Okay. So thing is the way you are blending your data will change the scope of your analysis. So you have to be a bit careful for that as well. There is arrival airport does departure airport and we have a single lookup list for all different. Airport if you blend it from a rival, it will show you arriving airport. If you blend it from departure will show you departure a point. We can blend more than one data source, of course, let me show it practically to you. So here it is Airlines dashboard. It's going to take a bit of time to open up. So I'm just going to continue with the session and then we'll come back to it and I'll show you this example where I've Blended one data source more than once. There's just one data source, and I have brought it twice in this dashboard.

Okay, so in the meantime Well, it's opening up. Let's proceed ahead with charts are widely used on on what particular situation is best to use them so bar graph line chart pie chart scatter plot histogram box plot. These are some of the widely used charts and when to use them anyone here from my batch. I'm sure they will be able to answer this question pretty well. I'm not sure about the others but do you have some understanding when to use a bar chart versus when to use a pie chart versus when to use line chart? Let me give you some context bar chart you use when you have a categorical data and numeric data may be or if you want to show count of categorical data. So when you are trying to compare categorical data, Maybe male salary versus female salary. Number of male employees versus number of female employees. We have categorical data show.

If you are trying to compare categorical data you use bar graph line graph is used to show Trend over a period of time. If you want to see how things are moving with respect to time is your sales increasing is your sales decreasing with respect to time. Mostly you use line chart pie chart is used to show the percentage contribution mostly in most of the cases you can Use it exceptionally also but most of the cases pie chart is used to show contribution percentage contribution to hundred percent of different components. Okay. So for example, you have your employee base what percentage of employer male what percentage of employer female don't use pie chart if you have more than three or four categories, okay, it's better best to use pie chart only when you have to at most three categories if you have beyond that, it's better to Use bar chart then we have scatter plot scatter plot is used to show the relationship between two continuous numeric variable when you want to explore the relationship between two continuous numeric variable you use a scatter plot.

Okay. So for example salary of an employee versus age of the employee. Okay, you want to see if older employees are getting paid more than younger employees or maybe experience of the employee versus the salary is getting These kind of relationship if you want to investigate you use a scatter plot. Then histogram is used to show the distribution of a continuous numeric variable. So for example, you have weight of different individuals how how many people are between way between 50 to 60 kg how many people weigh between 60 to 70 kg? Maybe if you are doing analysis for an Airlines and and you have some sample data you would like to analyze how much how many passengers or how much? Baggage weight.

Maybe you should provide to your passengers histogram will probably help you most of the people carry between 0 20 gauge is to 30 kgs of weight with them. It will help you analyze. Maybe you can charge some premium that histogram is used to show the distribution of continuous numeric variable whether employees on average they are earning on higher side of their earning all Louis egged. Then we have boxplot box plot again shows distribution of continuous numeric variable, but mostly used for exploration. It will also highlight the outliers for you. Okay, so it will point out each and every individual data point on a particular box plot, you will be actually able to see what is the distribution of a data. It's not a very common graph mostly used for exploration because you know general public they do not generally understand boxplot you have to take time and For to explain them what a box plot actually means so mostly it is used internally between analyst for exploration purposes how to be an expert with Tableau formulas here.

I have provided a link. This is like a reference guide for you know online Tableau help online reference guide you can follow this. Let me just open it up and show it contains details of different functions which are available within Tableau and with example certain examples. It's not very detailed. As such you know, each of the function is not explained through rigorous examples as such but it's a good reference. So we have numeric functions string function date function. Let's explore date functions within date functions. We have date diff data add example here did this example all these things are there if you want a more, you know tutor like approach where you want to sort of like get some context and those kind of things.

Maybe there are some good books. Written in Excel for Excel formulas which are very comprehensive contains exhaustive information about Excel formulas many of these Excel formulas are relevant to Tableau. Okay. There's one book. I would like to recommend formulas Excel 2013 formulas by John walking bash. We're pretty famous book and you just need to read this part but using functions in your formulas tell you text manipulation function date and time function counting some function select lookup functions are Are available in Tableau to some extent not a great extent.

Okay, LOD calculations. We are going to talk about LOD calculations. They are not going to be described here in Excel because this is a particular feature of Tableau. It's not a feature of Excel and it's a pretty recent feature LOD functions. They it was introduced in Version 9 only so not a very famous a such feature of tableau. Next topic moving on. What are the different workarounds? We use interval. Actually, there are many different workarounds some we have already discussed in performance optimization and blending, you know, like for example, you can use one data source more than once in a dashboard here.

Let's see. Maybe you're here the data is still being downloaded. It's going to take some time. So I'll come back to it so you can actually blend your data more than once in a dashboard then I talked about different. It's optimization instead of using IF else. You can use Simple calculation logic like you know comparison sales is less than 10 and then you can use Alias to show it in a meaningful way.

So there's so many different things available. You can create charts which are not available by default in Tableau, like hundred percent stacked bar chart few people are not aware of these tricks. So I'm just telling them to you here. There is a hundred percent stacked bar chart. Let me explain hundred percent stacked bar chart. I'm not going to tell you how it is created, but you can create this interview and you might be knowing it already. Maybe you do not realize it or recognize it by its name, but you might be you might have seen this chart already somewhere maybe interview itself. It's pretty easy to create in Tableau a chart which looks something like this.

This is hundred percent stacked bar chart, then we have donut charts. Now. This chart hundred percent stacked bar chart is not available by default. Okay, so you have to take certain steps to generate this. We have donut charts again. It's not available in Tableau donut chart. It looks something like this. Okay, all these charts can be created in Tableau. And there are workarounds available. You can create infographics what I infographics infographics. Like these Jazzy Graphics which are published in magazines and all these are infographics.

All right, these are infographics. Maybe they'll be some percentages and some numeric values shown infographics here might be here. You can see 47 125. These are infographics. You can create them in Tableau. If you want to you can overlay other charts over your geospatial map. Have you ever created a geospatial map which has pie charts? On top of it geospatial map having pie charts on top of it. Let's see if my data opened up already here. You can see just coming back to the previous topic as you can see.

I have Blended this airport look up data twice my original data it contains airport code. Okay, and then I have created a look up. So what essentially I have done is I have created a chart geospatial map showcasing departing flights number of Flights departing for app from a particular location and number of flights arriving at a particular location geographical information is available only in this lookup table not in your code data as you can see I blended the same data twice once for arrival once for departure, okay, and you can also overlay other charts over your geospatial map. So for example, you can create a pie chart on your geospatial map. You can do that just in case if you are not aware already. Let me show you an example of that again. These can be made available to you through my class recording.

So just request the support team to get access to my class recording and in case if you're not already aware of it, okay, and you would be able to see how to create these kind of geospatial map. Anyone who's not aware how to create these kind of geospatial map. You can see pie charts overlaid on geographical Maps Okay, so these kind of things are possible so there. This list can actually be very long and there are several different workarounds. Just keep your eye open keep your interest in tact and Tableau and you will learn a lot of work around their wonderful thing people have done one place where I would suggest you can explore these workaround is Tableau public server.

So people publish their work on Tableau public server, and they have done wonderful job. I mean magical things using Tableau. So just try and expose some of those examples they have published. T' try to recreate them from your side. Okay level of details your what are LOD when we will use them. Okay, elodie's level of detail expressions are recently introduced featuring Sky Blu 9.0. And before they were introduced there was this nagging error message you will get when you will try to mix up an aggregate function and an on aggregate function. So for example, you have different employs. Different employees and you have these employees are earning something. Okay, they are getting some salary you want to see how far away they are from the average salary of all those hundred employees. So you want to calculate average salary of those hundred employees and you want to calculate each employee how far away he is from that average salary? Okay.

Now it's a combination of row level function and aggregate function average salary is aggregate function individual salary is a role every when you want to calculate something like this salary of an individual – average salary of the entire hundred people group. If you were going to try this in an earlier version of Tableau, you will get this error cannot mix Aggregate and non aggregate arguments with this function level of details are a way to address this problem. Okay, and it's a big topic and needs lot of discussion due to lack of time. We won't be able to take it up so you can read more about them here. It's pretty simple straightforward. No rocket science. There is this syntax, which you need to be aware of level of details expression expression syntax fixed include exclude. Okay. So using fixed you can talk about I'm sorry using include you can talk about dimensions and exclude level of details expression X Plus, even you know, to be honest enough. I haven't used this feature much. Okay, because I have recently migrated to Tableau 9.00 a couple of months back and never really required to use this function, but Seems fairly simple and straightforward and there are certain examples given in fact for simple straightforward calculations.

You can aggregate the function like this, I think so. Yeah, here it is. So this is not going to work but this one is going to okay. So this has level of details expression embedded So within curly braces that we could just go through this article. I've provided the link in here and you'll be able to See more details on level of details of this particular error message was pretty nagging. Let me tell you A lot of times. I've tried to do things which didn't work out because of this particular limitation of Tableau and level of detail seems to be a pretty exciting feature to me and I'm pretty sure you're going to encounter it frequently extracts.

What are extracts extracts will increase the performance of your dashboard. They it's a way in which table you will extract the data keep it. Side optimize it for faster performance. There are certain formulas additional formulas which becomes available when you create an extract. So for example count D formula, there is a typo here. It should be count. Di not counted formula should be County formula and I'm talking about earlier versions of Tableau. I'm not sure if it's available by default interview 9.0 and above but County formula was not available by default unless and until you are creating an extract this distinct count for me. Le was not available to you only for extracted data and you can control Refresh on ever-changing data source, if you are using extract, okay, if you if you make a live connection data changes your Tableau report changes.

Okay, if you have scheduled refresh in extract, you can control the refresh allow users to refresh the data based on their preference or you can schedule it also if you want to so this lot of control which you get on data refresh by using extract Good feature, especially improves the performance how to refresh your visualization from your desktop environment. So go to data, okay here let's say if you have this data source go to extract and click on refresh. If you have created an extract if you have live data, you can refresh it right from here. What is data visualization? Okay. Now we are coming to common interview problems. These problems have been asked by you question. Number two question number 12 have been asked by The Learners. Okay. Now some common interview questions. What is data visualization data visualization refers to the technique used to organize and present information intuitively using different visual techniques, and it enables you to quickly answer this question and your data becomes a competitive Advantage instead of an underutilized asset.

If you're going to show your data in tabular format, it will be very concerned, you know. Confusing for the users to digest that information if you show it visually in form of charts and graphs it becomes visually appealing more user-friendly. So there are different techniques of data visualization which involves things like when to use a particular kind of chart and what kind of situation what kind of color themes to follow some of those things. We already discussed in some of the previous questions. Okay. So all these encompasses the field of data visualization making information relevant connected to each other enabling quick slicing and dicing of data filtering of data pivoting of data giving users an option to pass on feed inputs through parameters. All these are part of data visualization techniques. What is the difference between quick filter normal filter and context filter quick filter is like when you right-click on your visualization, you know, you create a quick filter like I've done here. These are quick filters. These are normal filters.

For example Department name in your if you want to edit this filter, you can go and choose wildcard condition some condition top and numbers all these things you can do. This is normal filter. These are quick filters. So this is again a common question people may ask you what is normal filter was a what is quilt and then we have context filter context filter is when so what happens in Tableau is when you are applying normal filter or quick quick filter, they are going to be independent of each other. Let's say you have again HR data set you are going to filter. Data based on gender. Okay. So you have applied a filter on female employees.

Then you want to filter data based on geographical location. So let's say you have applied a filter on India. So what is Tableau going to do table is going to apply filter on female employees. And then again, it's going to apply filter on all the data set, you know, the end all the rules on India and then it will show you that data.

There are some drawbacks. Backs of it. So what happens is let me show you an example on this HR data set itself. So I am going to pull up Department name and six salary and I will show let's say I have a salary and I'm going to take gender. supplier filter of female here And Department name top, it's empty king top three values by fixed salary average salary as you apply, okay? Let me just remove the gender.

I'm really not able to think of a proper example in here. So what actually happens is when you apply normal filter. They start from scratch for each filter you applying. Okay? Let me see if I can simulate that thing in you. Let's see. We have chosen both male and female. So top three departments eeo office HR and testing okay if we apply a filter of female. Okay, then again, you can see see you office HR finance and facility and testing. Okay. What if we take this gender and keep it here? We have to put a context a actually it's not going to be visible until we put a context where the so what context filter does is it will apply a filter by default.

Okay, so you can restrict Department name here. And then if you pull in gender, let's see the results actually changes. It's not actually changing in this particular case. I've chosen a lame example. Okay, so context filter what it does is it will apply a filter at first level and then rest of the filters which you are applying will be applied later on. Okay. So the next set of filters will actually work on filtered data. So when you apply a context filter a new data set a temporary table will be created and then your rest of the filters will work on that temporary table. Okay. Going to speed it up your thing speeding up your dashboard. And also when you are using certain situation like top 10 in my particular example, it didn't work because that was not really relevant.

I'll pick up some data example and probably post it as a article on the Tableau this blog section. Okay of a Eureka and there you can see mostly for top and results. We use this context filter. So we have normal filter. We have quick filter. And we have context filter context filter will create a temporary table and then other filters which are going to apply will get applied on the context filter. So it will be level filter one filter applied. Second filter will get applied to the filter data not all the data usually used for top n values extracting top n values these kind of things.

Okay, what is data blending? We have already discussed it. So I'm not going to go back to this discussion again. And when do we use data? Learning so data blending is used when we are pulling in data from different different sources. For example, we have some Excel worksheet. We have some table from Oracle database. We have some table from Tableau server. So in that case we can combine them all into a single View using data blending but given a choice. You should always go for data joining if your data is coming from the same data source, let's say you have one Excel file which has multiple tabs and and you have an option of either blending the data or joining the data go for data joining.

If you have an option, it makes things more intuitive straightforward and all data blending is generally used when you are working with different data sources another important question. This one is quite important actually. So what are the differences between TW B LT W BX? So twb is General Tableau file and TW biessed is Tableau packaged workbook. So twb Tableau workbook and TWP. This Tableau packaged workbook W packaged workbook is like a complete set of data plus any image, which you have kept in your dashboard. All of them combined together into one.

Zip file T. WB W workbook is just a set of instructions which table utilizes to draw your visualization. Okay, it doesn't contain any data if you have used a picture if you have been embedded a picture that is not going to be available. In fact, you can open up a twb document by right-clicking and opening it up in a you know, if you have notepad plus plus you can open up the twb and see the code behind it. It just just contains the instruction there is no data. So how table you should repaint your date on screen those instruction. That's it. What are the difference between groups and set difference between group and set can be a bit confusing for some people actually, you know, you can improvise their functionality and Make them behave similar to each other. So but essentially groups is something when you club members individual members together into a group set on the other hand will create an in-and-out sort of a feature for you.

So for example, we have job title. I can go and create a group here. I can say co-manager senior manager SVP and VP. These people are People manager. Okay, and these people are the people associate Junior associate senior associate SME support staff. These are individual performers. I can do that. Okay, I can say okay. This is how we create a group then we have set. Okay instead what we do is we create a set so we'll say create and then we create a set.

Okay, and then we can apply a condition. Let's say You want to create a set of all those people whose average fixed salary is may be greater than let's say 10. Okay, and you you can say this is a high learning people. So this is how you create a set. Okay, and then you once you create a set then what happens is you can pull the set into your analysis and maybe you know pull in some other details maybe and then you can see this in and out. This is one way or you can use it as a filter so you can just right click here and you can choose this option show member inside. So what will happen only those people where those cases where the learning is high will show up so here for each of them.

The salary is going to be greater than 10. Okay, as you can see so it can act as a filter on the fly. If you choose this option Show members at if you choose this option, you'll see how many people are lying in that range. How many people are like be on that train? So basically they serve different purposes, but you can improvise them use them along with formula and make them behave like each other.

So for example, I can create a group and then I can create a formula on this offense formula sort of and then I can use it simultaneously to create like a set like feature also, you can create unions and intersection inserts. I hope you are aware of all those things so you can create a union of sets you can create. The intersection of set all these things you can do. So these are differences page shelf. What is spatial you can analyze data on Tableau using a feature called page shelf by shelf will actually create different pages of your different situation. So for example, if you have I'll show this actually I'll show page shelf in action to you.

You can create animation like feature using page shells. You know, you can see different department and their head count Trend over past five years. So pmo versus testing was a CEO of is versus development versus HR all these different things are appearing. So what actually happened was I put this information in page section. Let me just open up this sheet for you. I put this department name in the page section and then Tableau created separate pages in telling together for each of the department and then you can play them together and create this animation like feature.

These are also called as motion charts in case you may hear different versions of this question. What is spatial what our motion charts and how you can create animation like features interview. So for all of them answer Remains the Same and here is a detailed usage guide for page shelf explain. When would you use join versus blending in Tableau again? You have the same question. I've already told this to you blend is used when you are getting data from different data sources. Okay, and joining is when you are getting data from same single source. So for example all data coming from same Excel file all data coming from CSV files you use joins one data is coming from CSV.

Another data is coming from Excel. Another data is coming from Oracle you use blending? Okay, if given a choice always go for join it makes more sense. Or intuitive everything in one single place. No need to enable the link anything like that. No need to Define primary calculations. I can recalculate these things are not required. When you use a joint does that make sense? We have talked elaborately about joins and blending multiple times. We have I treated this. It's an important question. That's why okay, what is default data blending join? I've already talked about this data blending is the ability to bring data from multiple data sources into one Tableau View. That is what data blending is in, you know people from database background. They tend to think of data blending as like join data blending is not actually a join.

Okay Tableau picks up data based on primary data source, it looks at primary data source, then it looks at secondary data source and tries to match them and pixel pick up data that way but it's not actually a join which is going on behind the scene. So there is A bit if you try to think of it logically or you know data blending is not actually join. No additional column is created. Okay. So when you do a join new columns are created your data changes sort of okay in blending. Your data doesn't change. It's just a method of bringing fields from different data sources together in one single view. That is what data joining is. I don't know how it works behind. The scene but how data blending actually you can utilize in different Innovative things different innovative ways. Okay, so sales, you know that you can actually capture date of the sales and you can compare the monthly headcount from a different data source, and you can blend them one section. You can aggregate the date into month. Another section already has month. You can blend it one calculated field one course.

Field okay, you can blend them together. No problem at all blending you can do in multiple different ways. You're just limited by your creativity at times it can be nagging I do agree. So at times you may encounter problems, which doesn't make logical sense, but then it's just the way you have defined your blending. You can if you change the scope you can make it work as well. I've done pretty crazy branding and I've seen people doing crazy kind of blending and creating wonderful results blending is a powerful powerful feature, but given a choice go For data joining it's easier and intuitive also and you can emulate left join right join inner join by setting your primary and secondary data source the way I've explained to you already.

So if you choose your primary data source and secondary data source will be a left join. If you flip them. It will be a right join isn't what you're choosing as your primary and secondary if you filter out null values, it will become an inner join. Okay, there is no option. Of creating a full outer join. What do you understand by Blended access in Tableau measures can share a single axis so that all Max are shown in a single pane instead of adding rows and column to The View when you blend measure there is a single row or column.

Let's see this one. I think we are talking about dual axis in here. Oh blending of Access compare. Okay, so essentially, you know when you have two different measures, Can actually blend their axis. It's like let me see if I can create an example for you we have. Number of records and we have date of joining. Let's say and we have created.

A trend line and let's see we have different department names in here. So let's see. I mean essentially if you blend the access everything will come in one single way. Actually it stands for. When you have two measures not two Dimensions, but two different measures so we have this and we have this fixed salary. So I'll take fixed salary and put it in here. You can then you can blend the axis. Are they talking about dual access in here might be I think that's what they are doing.

This is blending of accesses what they are saying showing okay and then you can sync them if you want to so we have two accesses here if you want you can sync them. So Let me see if I can sing them together. Okay, I'll get back to it. I haven't used will access for a while. Now. There is an option here through which you can sing the axis actually, okay, and that will create the same access levels in both the fields. So essentially one line will be on top another line will Crawl Through the bottom. So yeah right side. We do have this option. I'm not able to see it. You're maybe going to because of scope of my calculation or some some problem synchronized dual axis option is there. Where do I don't know? Oh, I have to enable dual axis. First. Is that dual axis enabled? And okay, so I have to remove include 0 sorry from here. And then right now this option is not enabled for some reason uniform access may be independent fixed.

It's not working right now synchronizing synchronized will access is not working for some reason maybe because of scope of my calculation. I'll check this. Okay why this is happening. I apologize for this one. This option is not actually very difficult to use for some reason. It's not working right now. Okay, what is story in Tableau? This is again a new feature. I think it was introduced in version 8 if I'm not wrong and story is a way to present your findings or your analysis in a step-by-step manner. That's it. I mean Story Probably is a fancy name. It's nothing new. So basically you create your worksheet and your dashboard story is a way to put them in sequence so you can give your users a guide. That tour. Okay, like if you trying to present something some findings interesting findings to you user. You can first build the context. Let's say we're talking about HR dashboard. Okay chart Workforce analysis. We have done of a company first point you would like to say this is the workforce profile of our company B's many male employees.

Are there these many female employees are there. These are different employees working in these different grades. So you will create your first story line, which is This is our HR profile Second Story point you will compare the ratio of male and female and you will say there seems to be a gender bias. If there is a gender bias. Okay, if you are investigating on gender bias third story point you can see where this gender bias is coming from. Okay, so you will say okay. These are different departments where the gender bias is the highest. Okay, then third fourth story Point what what might be the reason you can provide some hyperlinks some external sources.

For the reason then fifth Story point maybe you can show the revenue stream of those Department which have healthy gender mix and probably they're better. Okay absenteeism rate is probably less in those department where we have healthy gender mix and then you can put your final point that you know, we should have healthy gender mix so story is nothing but a sequence of worksheet and dashboard to convey information in a more meaningful manner. Okay. It just gives your uses a guided tour. They don't have to hop for information. One place to another without knowing anything you guide them. It's like those slide shows which are available on Yahoo. And you know different websites where you click on that arrow and it takes you to the next picture maybe some Bollywood story or some you know, some fancy story.

They want to share with you. Similarly. You can create a story in job you as well make sense. So these are story points. And what is the difference between discrete and continuous interview? It's a actually consistent difference. It's like part of all round analysis not just in terms of Tableau. What is discrete and what is continuous discrete is something which has limited number of values discrete values are counted as distinct and separate and can take individual values within a range. So for example number of Trades in a sheet customer name or row ID or state? Okay gender for example coming to my HR example gender or departments. These are discrete values. Okay, if if you are talking about numeric values, I mean, these are actually categorical values to be specific discrete and continuous is more when it comes to numeric values. Okay. So discreet is something which has limited.

Were of values for example your performance rating. Let's say your performance rating is given in value of 1 2 3 4 5 where one is it's a the highest rating and five is the lowest rating. This becomes discrete data. Okay A person can get a rating of one or two or three or four or five continuous data on the other hand takes continuous value like it can contain any value. It can have any values for example the age of a person Can be 25.3 two years, you know height of a person can be 5 feet 6 inches. Okay, it can take any value. So continuous versus discrete. These are common terms in analysis general terms and then we have categorical values as well. Go categorical versus numeric values measures and dimensions for these kind of things.

So discreet values which can take stepwise values continuous values, which can take any values. How do I automate report using Tableau Software to automate the report do your work and create your dashboard publish it on to W server. You will have the option of scheduling the report you schedule the report when you want to get it refreshed when the users will open up the report. So Tableau server will automatically run the job for you based on the time which you have provided to them.

So let's say month start first date of the month. So 12 o'clock, whenever the date changes its going to fetch the data refresh your visualization save it users will come next day open up. The report will be able to see new information every month start. They'll be able to see that. This is how you can automate it. How can we combine database and flat date file data in Tableau desktop, of course through data blending. So you pull data from your table and flat file and then go to data edit relationship if you want to so, for example here, I have done some data blending. I've got two data sources here. If you remember two data sources this in this I go to data edit relationship one data source to Second data source, you can define a custom relationship. Also if you want to or let it be Custom I do not want my blending to be done on rating.

I just want it to be done on employee ID and then say okay. I'm sorry this sheet 13. Nita and it relationship this is what you want. So this relationship has been defined now and then now you can pull in data from these two different data sources and you will be able to blend them successfully. So these steps you need to take if you have two data sources, they have common column names you do not need to do it this data and edit relationship is not required if they have common column names and if that's what you want to blend the data on then you do not need.

Go to data and relationship that's not required. But if the name is different, then explicitly you have to Define this so just be careful about this one and then next step you go pull data from one data source, go to your second data source, you'll be able to see this chain like link enable it and then perform your analysis is drag the fields which you want to then go back and you can take employee count whatever, you know, so you can perform you on this is this is how blending work? X what are the platforms Tableau server can run on Tableau server can run on Windows and Mac.

I don't know if this might be a very relevant question for a tableau developer. But yeah, okay. How do you publish table reports to Tableau server to publish a report go to server choose this public publish workbook option. It will ask you for your credentials and you have to provide your credentials. You need to have access to a tableau server. Okay. So first you need to probably click on the sign in. Provide your credentials then choose publish workbook. It's very simple go to server sign-in provide. Your credential published workbook will give you the name of the workbook, which you have saved by you can change this name. If you want to you can provide, you know user permission as well. Let's say you have 10 different sheets and you want to publish only few of them. Just check them. Whichever you want to publish and check the one you do not want to publish it will still be part of your Tableau desktop file, but it won't be published to the So, okay, and then you say publish and it's going to work you can include external file show selection these options you can enable it's very simple all menu-driven very easy thing Define parameters in Tableau, and they're working Tableau parameters are defined variables values what our parameters parameters are away to accept inputs from your user.

Okay and let them interact with your dashboard. Okay, and you can place parameter controls. Want your dashboard users can pass on values through those parameters you can use those values in calculation and filters here is an example. You can create a calculated field value returning true when the score is greater than 80 and otherwise false using parameter one can replace the constant value of 80 and control it. Dynamically dynamically in the formula. Okay, so it's parameters are a way to take user control. Okay. So these are parameters. What are the difference between Tableau desktop? Tableau server Tableau desktop helps you create data visualization workbook creation.

These kind of things it will help you do Tableau server is used to distribute these workbook to the end-user. These workbook can be interactive. You can enable your users to apply filters slice and dice the data pass value through parameters, and all of these functionality will be available through Tableau server. It will be available just in case if you do not have access to Tableau server. Does Tableau desktop also become redundant to you? No, not at all. There is Tableau reader which you can still use for offline viewing of the file. So you can ask your end users to install Tableau reader. It doesn't involve any cost. You can create your work using Tableau desktop, pass it on and they can use Tableau reader like a PDF reader. Okay, so we have acrobat files these Adobe Acrobat files. There are software to create those files and there are software's which are supposed to just read those files tab you. Dinner is such a software. It can read your dashboard. All the interactive features will still be intact. It's just that the user won't be able to create anything. No new charts.

They won't be able to create a formula or something, but they can apply the filters quick filters. They can use they can click on chart and you know slice and dice the data all those things will still work anyone here who wants to learn about hundred percent stacked bar. Let me just quickly show it to you. It's pretty simple so stacked bar. Let me Quickly show you how to create a hundred percent stacked bar. I'm going to show you male versus female ratio the thing which we talked about number of Records this data. This is HR data set we have different departments employees working on those departments and we have male and female employees gender demarcation.

I've created a quick pivot table like summary table. So we have CEO of is and development HR Finance male-female all these things we have, okay. Let's create it step step by step and I'm going to show you how a stacked bar plot is created a hundred percent standby power so different departments and number of employees working in those Department. I'm going to create a bar plot. First of all. And then I'm going to flip it. So, you know, it's up to you if you want to flip it or not. So CEO of his development HR Finance [ __ ] testing and then I'm going to take gender through it in color. So we have male and female employees working in each of these departments. I want these bars to be of equal height and I want to show values as percentages. Okay, so I'm going to take number of Records. I'm going to throw it in here. These are male employees working in development.

These are female employees working in HR and I'm going to click here quick table calculation show it as percentage of total, but then what has happened this is actually percentage of entire total. I want this hundred this particular bar to be a hundred percent. So these two should add up to a hundred percent these two should add up to a hundred percent right now. It's not happening. Why because the scope of calculation is like this where everything if you add up Get the will become hundred percent. I want to edit that table calculation. So I clicked on this drop-down. Choose edit table calculation table across I'm going to choose table down instead of table across a apply. Okay. Now each of the bar is adding up to a hundred percent, which is what I wanted. Okay. Now these different bars broken up so each bar is now adding up to a hundred percent, but I do not want to showcase that Co offices like, you know tiny and this development is biggest. I want bass to be of the same size.

Because what can I do? I can click here now. And this is a work around this is not straightforward and intuitive and that's why I kept it in the workaround section. I'm going to go to this section. What is this section defining? It is defining this axis. That's why I have to go you I have to set the axis now. I set the values I have to set the axis. Quick table calculation. I'm going to say it as percentage of total. Okay, and then again Same step. I'm going to perform table across table down. I'm going to choose say apply. Okay, this is my hundred percent stacked bar chart showing clearly that CEO of is has probably bad whatever gender ratio. This one is good HR finance and facility this I was not able to interpret properly here.

Why because this bar was showing up as like small and all but if I convert this into a hundred percent Stark bar I can see this portion is biggest for HR. This has good gender ratio this how you create a hundred percent start barcode will move on and we'll take a look at the success stories of Tableau. So almost every company uses tablets that business intelligence tool and all of this company that you can see in front of you in the screen.

They all use Tableau starting from US Air Force to Burger King's Citibank, you can see these are all all different companies, but they all use tablet. These are a few of the comments made by the most influential people from a particular company. You can see Ryan Greener from Deloitte says that Tableau is changing the game for us. It's reduced the time that they have to spend on Lower value add activities similarly Gerard and namesake from one Kings Lane say is that it increases our sales it decreases our cost.

There is a direct impact it just gets you in. I'd Foster and you can read for the other two as well. So there are many success stories. People are absolutely loving Tableau and once you use it, I hope that you love it too because you can do and you can play around with data in different kind of ways with Tableau.

Thank you for watching this tutorial and I'll see you next time till then Happy learning. I hope you have enjoyed listening to this video. Please be kind enough to like it and you can comment any of your doubts and queries and we will reply them at the earliest do look out for more videos in our playlist And subscribe to Edureka channel to learn more. Happy learning.

.

test attribution text

Add Comment