Tableau software is a totally new approach to business intelligence It’s designed for anyone who has questions about data Tableau Desktop is a drag and drop product people at any skill level can use to visualize data on the fly, at ten times the speed of a traditional business intelligence program With Tableau Server, people can publish and share interactive dashboards and reports and then work with them on any web browser or on a mobile device Today I’m going to show you the power of these two tools by playing the part of a senior manager at a chain of retail stores In order to make my target profit goals this year I know I will need to make some changes, but i’m not sure yet what those changes should be I have a lot of questions about how my stores are performing Let’s begin in Tableau Desktop Tableau Desktop allows me to connect directly to any data sources you see here including others, through ODBC You can also connect directly to Microsoft Access excel or a .csv file In Tableau 8 you now have the option to do direct connectors to Google Analytics and Salesforce Unlike with other business intelligence tools people who have questions about data don’t have to rely on IT or complex scripting to connect to their database I simply enter the server name my user information, connect the database and select the tables I want to export In this case I’m going to connect to an Excel file that I know has the information I need about sales of my company Connecting to an Excel file is even easier than connecting to database You just navigate to your workbook, select the worksheets you want to use and click OK In this case I know that the order sheet contains the numbers about sales and profits that i’m going to need so I’m simply going to connect to that table Tableau is now giving me the option to connect to a live to my database or to import all the data This option is given whenever you connect to a dataset It allows you to connect to a high-performance database if you have one Or bring the data directly into Tableau’s fast data engine In this case I’m going to connect live Once I’m connected Tableau separates my data into dimensions and measures Dimensions are categorical fields Like customer name, customer segment, region and state Measures are numerical fields like discount, profit and sales Typically when you’re using a measure you’re either aggregating it, like a sum or an average Or you are doing a calculation with it In order to visualize my data, all I have to do is drag a field onto one of the Shelfs Each of these shelves has a different purpose For instance, the Column shelf corresponds to the x-axis of the visualization whereas the row shelf corresponds to the y-axis As you might expect, the Color shelf, Label shelf, and size shelf correspond to the color of the visualization, the labels that are on the visualization, and the size of the marks on the visualization The first analysis I’m going to do today is look at my sales data over time, as you see here I want to understand if there’s a cyclicality or seasonality in our sales, and what can be done to alleviate it We can see sales have gone down between 2009 and 2011 but they’re on the uptick for 2012 One powerful feature of tableau is the ability to work flexibly with dates I can specify many different options for how I’d like to see and utilize my dates I can drill down to quarter, month, week, or day — as far as I want — and then, using the Undo button, I can go back to the level of detail Dates are a special type of field, so we treat them specially. We don’t expect you to have to do a lot of work formatting your dates With Tableau, you bring the dates in, and use them exactly as you’d expect to Part of this date flexibility is the ability to switch from a continuous view of dates, like the one you see here, to a discrete view of dates, like this one Here, we’re summing the entire data set for January in all years We’re doing the same for February,

and so on This view is particularly useful for the analysis I’m trying to do because I’m trying to understand the cyclicality of my data As we might expect from a retailer, Q4 is particularly strong May through July, however, are comparatively weak As a manager, I want to know if there are particular product categories behind this cyclicality To answer my question, I want to view this same data set by product category To do this all I just drag Product Category onto the color shelf Tableau understands I’m trying to separate my view by the 3 categories I use in my data set Now in my view I can see the sales figures — and the cyclicality — for furniture, office supplies, and technology I can see that furniture has a pretty high cyclicality, and may be contributing to the dip we see between May and August Now we’re looking at aggregated data for all years here so I want to confirm that furniture’s cyclicality holds true across all the years, and that there’s not a single year at the root of the problem When I highlight Furniture, I can see that the cyclicality is pretty consistent across all years, with upticks in the 4th quarter and a relative lull in the summer This analysis will be useful in deciding how to stock my shelves in order to minimize the bottom-line impact of this cyclicality I want to make some changes to this to you before you share with others For instance, I’d like to change the month names so that just the first letter of the month is displayed Tableau makes this easy to do While Tableau makes it easy to explore data visually, so with just a couple clicks, I can duplicate all this information as a crosstab and share it that way, too I’m going to rename this sheet “Crosstab” and rename the original “Trends over Time” Before I move on to my next analysis, I want to make a change to my metadata layer In Tableau, I can change my metadata however I want, and it won’t write back to the database This is great, because it allows me to change and view my data exactly as I want to, but I don’t have to worry about causing problems in any production systems Here you can see I have Order ID as a measure Even though it is a number, it’s not really a measure For instance I’d never take the sum of multiple order IDs or take the average over a group of them It’s categorical in nature so I’m going to drag it up into Dimensions Again I haven’t changed anything in the production database I’ve just told Tableau to handle this piece of information differently I’d also like to combine Product Name, Product Category and Product Sub-Category into a hierarchy, called “Product” so I can more easily see patterns Some cubes include this functionality, but in Tableau, we give it to you with any database you connect to I can also change the alias of any individual item within my fields For instance, I can rename “Technology” to “Technology and Peripherals”, and that will update this item’s name whenever I use that field in Tableau Again I’ve not impacted the underlying data base Now that answered my question about cyclicality I’d like to ask a question about geography Are our least profitable stores concentrated in a certain state or region? Geography is a special kind of data, so we treat geographical information in a special way in Tableau However, you can use these fields the same way you’d use any other type of field — by dragging and dropping. Tableau’s built-in geo-coding allows you to visualize geographic data immediately, without tracking down the latitude and longitude data yourself Tableau automatically recognizes the names of names and coordinates of countries, states, counties, cities … even congressional districts There’s a lot you can do with maps in Tableau For instance When I drag out a measure, like profit, onto the map, Tableau automatically turns it into a filled map, indicating by color the profitability of a state

This is possible at all area levels of geocoding, so I could look at it by country or city, even a congressional district as well When I drag out a second measure, like Sales, Tableau changes the visualization so that it can represent both pieces of information The size of the circles, or marks, represents the Sales in that state, while the color represents Profit To make this a little bit clearer I’m going to increase the size of the marks and also add a border That’s better It looks like, over all, most states are doing quite well, but we can see it right away in Montana has negative profit we’re going to try to tell and even more useful story and changing his color palette as you can see I’m going to try to tell an even more useful story by changing my color palette a little bit As you can see, there are lots of fast and easy ways to control your color palette in Tableau Seeing the profitability at the state level is helpful, but I’m guessing I might find more unprofitable stores if I look at a lower level of detail I’m going to bring zip code onto my level of detail, so that I can visualize this information at that level Although this default view is good, I can change, modify, zoom and edit this view however I want, with just a couple of clicks Right away, I spot zip codes in Vermont, New York, and Maine that are exceedingly unprofitable I can hover over these zip codes and get more information in the tooltip We’ll dig deeper into these problems as we get further in the analysis I’m going to zoom out and rename this sheet “Profit by Zip Code,” and move on with my analysis Now that I’ve looked at trends over time and profit by zip code, I have questions about how my stores are performing by product category and sub-category Sometimes when you’re creating a visualization you’ll know which visualizations you care about, but you don’t necessarily know how you want to display them For these situations, Tableau includes Show Me Show Me is a tool that helps you select a visualization type based on fields you select So, if I select Product Category, and then use the Control key to select Profit as well Show Me automatically highlights the visualization types that will work well with this data So with this data, I could use a pie chart, a bar chart, a bubble chart, a treemap or any other highlighted visualization type But i’m going to choose the bar chart I used Show Me to create this visualization, but now I’m free to edit it however I want Now I’d like to look at this information by region to see if there’s pattern I see right away that furniture in the East is a particular problem But I wonder if it’s furniture generally, or if there’s a specific type of product driving this problem Because I created the category hierarchy earlier, I can click this plus button and drill down, automatically expanding the view Now it’s clear — tables are driving the profit loss in the East In fact, tables aren’t doing us much good anywhere I can make this even clearer by dragging Region down onto rows And I can make this view more compact by hiding the header for product category All of these features, from Show Me to hierarchies to the easy formatting changes, allow you to visually identify outliers and problem points in your data, in ways that may not be obvious in a spreadsheet I’m going to rename this sheet “Profit by Category” and move on to my next analysis I have questions about Cost of Goods but, if I look at my Measures, the data I brought in doesn’t include that information Fortunately, Tableau makes it easy to bring in a second data source and combine it with the first In this case and going to connect to my Cost of Goods Sold data, which is stored in a csv

When I bring the data in, it appears right alongside my original data set The first thing I want to look at is the relationship between Cost of Goods Sold and Profit When I view my COGS data set, I see that Tableau has automatically linked the two data sets on Product Category, and I can join on that field When I drag out Cost of Goods sold, I’ll be able to see that data for all my different product categories I’m going to turn this into a single scatter chart, and then drill down This view is interesting, but I want to switch my axes, so I can see if the COGS has an effect on profit It’s hard to tell still so I’m going to add a trend line It looks like there’s a slight downward trend — profit decreases as cost of goods sold increases I suspect there’s some variation in this trend by region, so I’m going to consider region information in my analysis by dragging region onto color Alright now it’s clear This downward trend is much more pronounced in East And in the Central and South we actually see a positive trend Without writing any code or custom scripts, I’ve brought in a secondary data source and added a really valuable layer of analysis to my visualization I’ll rename this sheet “Cost of Goods Sold” and now we’ll go ahead and create a dashboard A dashboard in Tableau is a way to bring in all the different sheets you’ve created, and make them work together in a single analytical view I can simply drag and drop to bring any of my sheets out onto my visualization When I bring out a sheet, Tableau shows me exactly where it’s going to land Since my data is on an annual schedule, it would be nice if I could quickly filter the data in my view by year In Tableau, we allow you to enable this kind of interactivity with a small widget called a Quick Filter Once I create a Quick Filter, I can select from a number of ways to interact with that filter, including single value and multiple value selection types Now I can go back to my dashboard, and show that filter here At first, the filter will apply only to the worksheet in which I created it, which is sometimes useful, but in this case I want it to apply to all the worksheets drawing from this data set, and I can easily do that If I wanted to, I could also choose specific worksheets to apply this filter to Now, all the worksheets on my dashboard update so they’re only showing data from 2012 Quick filters are powerful because they allow you to tell a big story in a compact space Instead of creating multiple reports for this data, quick filters allow any individual to drill down to the data that interests them Quick filters aren’t just available for categorical data I can also enable them for measures, like profit or sales Sometimes, however, you don’t want to include a widget on the dashboard to enable enable interactivity You might want people to make selections in one sheet, and have other sheets update based on that selection I can do that easily in Tableau, by selecting the Use as Filter option on any worksheet Now, whenever I make a selection in this sheet, the other sheets on my dashboard will update to reflect the filtered data Earlier, we were looking at a few stores in the northeast that were having problems with profitability Now, we can drill into the Profit by Category view, and learn more about the problem Again, it looks like tables in the east are the problem Creating a dashboard isn’t only about the interactivity It’s also about how you present the interactivity and enable it for use So often times, in presentations you want to use a title I’m going to rename this dashboard Profitability by Store Then I’ll turn on Show Title In Tableau, all your text is easy

to edit and format I can change the background color on the title or on the dashboard as a whole, add a border, or change the font. The same is true for my Quick Filters, or on an individual worksheet level I’d also like to move my date filter to the top of my dashboard, so it’s a little easier to see, and maybe change the title so that it’s a call to action When you create a dashboard in Tableau, you’re creating an interactive application that allows many people to view and ask questions about the information that’s relevant to them You don’t have to create multiple reports for different people and different questions In many cases, a single visualization can help lots of people answer the questions that are important to them Now that I’ve created this interactive application, I want to share it with people in my organization, so they can use it too As I was explaining earlier, Tableau Server can be installed on any Windows Network Server, so your own team can manage it and secure it. In fact, Tableau Server will support your existing security and data architecture You can cluster it to scale up to as many users or as much data as you need Now that I’ve created this interactive application, I want to share with people in my organization so they can use it too So now I’m going to name my workbook Profitability, and before I share it, I can control exactly who can see it and what permissions they have with it leveraging existing groups and permissions within my Windows authority Now, I’m going to publish my dashboard Once I’ve published, I can preview the view on Tableau Server And then open it in my own browser This view is interactive, just like it would be on Tableau Desktop I can now share this visualization with anybody who has a web browser, with no special equipment needed This browser interface works on any web browser, including a browser on an iPad or other tablet Other people can also view the visualization by logging into Tableau Server itself They’ll be able to view all the visualization that have been published, as long as they have the correct permissions Another nice feature of Tableau Server is the ability to filter down to a particular view within the dashboard — whichever view matters most to you — and have Tableau Server remember your changes So the next time you return to this dashboard, you’ll see the exact same view you saved before Because you logged into Server using your secure credentials, your choices only impact your view of the data Others inside your organization can select their own views If you’re using a database with your visualization, a nice feature is the ability to refresh that live connection within Tableau Server This way, if your data changes frequently, you know you’re always working with real-time data If you ever do need to change the view, you can simply republish over it using the same name You can also subscribe to a particular view using email The browser functionality in Tableau extends beyond just a laptop Your team can access their Tableau visualization from your secure server anywhere, like on this iPad

Here I’ve logged on securely with a web browser, but I could also use Tableau’s native application for iPad or Android I can filter, zoom or even select As you can see Tableau is native and beautiful on the iPad And I can explore this visualization the same way I explored it on my laptop In summary Tableau Desktop and Tableau Server are the complete business intelligence solution for anyone need answers to questions With Tableau Desktop and Server anyone can explore data with beautiful visualizations quickly and on the go