Over the next few weeks, we are going to learn how to use the visualization software tableau to both find and eventually communicate answers to business questions. Why do we choose tableau as the software? We think you should know by the time you finish this course. Well, let's hear from Ryan Mason, lead of Analytics Tools and Reporting at Square, the finance company who made those little square devices you can attach to your phone, so that anybody with a mobile device can accept credit card payments. >> Hi, I'm Ryan Mason, I work at Square, and I'm the lead of the Analytics tools and Reporting team. >> I asked Ryan what are the main software tools you use in your job as lead of analytics. This is what he said. >> So I'd say that the data that we work with in Analytics at Square primarily comes from databases, from a variety of different services. Ours live primarily in MySQL, although to access them we use both MySQL Hive and Vertica which is a high performance column store database. So interacting with those we have wrappers for writing queries in Ruby and Python that we use to kind of extract data and then interact with programatically. And then once we have data to work with we primary use Tableau, some people use Excel. The most widely used tool for analytics in the world. >> Ryan is one of a growing group of data enthusiasts who has made Tableau one of his go-to analytic platforms. That's because Tableau is an extremely powerful visualization tool that allows you to make visualizations of your data in seconds. Whereas other tools, like Excel, may require minutes, or even tens of minutes, to make the exact same visualizations. Further, one of the things that makes Tableau unique, is that when you make these visualizations, Tableau automatically uses best practices in visualisation science to format the graph view, so that you pay attention to the correct things. It can do this because the founders were world leaders in visualization computer science from Stanford. You may have noticed that the most recent versions of some more traditional analysis programs like Excel started incorporating more advanced visualisation features as well. Many say Tableau was the main motivation behind these changes. Tableau was so unique when it started that other companies had to adapt in order to stay competitive. Although other companies are trying to catch up, Tableau remains the leader in this area. Especially given my background in neuroscience, I really like this aspect of Tableau. One of the most important features in the professional version of Tableau is that it can connect to most kinds of databases and import data of insanely big sizes. Whereas Excel breaks after importing 1 million rows, Tableau, at least the professional version of it, is ready made for big data. In fact, under the hood, Tableau is actually running many of the same database queries you will learn how to run by hand in course four of this specialization. Let's hear how Ryan describes his use of Tableau. >> So I use Tableau a couple of times a week, and have for many years. For me, it's the fastest way to turn a table of data into a picture about what's going on. It allows me to very rapidly interact with it, to understand different slices in different views. Certainly I could construct all of the things that Tableau generates in SQL queries, but manually writing them is quite tedious. So what I like to do is, start with a fairly large data set, load it directly into Tableau, interact with it to try and figure out what's going on. And then maybe develop by hand some aggregate queries to monitor what's happening. Or maybe let Tableau do it depending on the scale. >> Another feature that Tableau is famous for is it's very easy to make dashboards. Much like dashboards in cars that allow drivers to monitor important things going on with the car. Business intelligence dashboards give you at a glance views of key business metrics or indicators real time as they're happening right now. Their goal is to allow anybody looking at them to immediately know whether something is going wrong or going right with the business. They're often designed and configured by an analyst, but then published to people all over the company so that everybody can keep an eye on them. Programs like Tableau allow for these dashboards to be interactive so that people all over the company can play around with them and get a feel for what the data will look like. Here's an example of what a business dashboard might look like. >> This is a sales dashboard from the Tableau website. The most important information is up here at the top where it will get the most attention. You can see the Total Sales, the Number of Deals, the Avg Deal Size and the Revenue per Salesperson. But, of course, there's a lot of other interesting information on this dashboard, as well. Over here, you can see the Revenue Over Time for each year, individually. You can also see the Running Total. And again, if you hover over it, it even gives you more data. Down here, you see the sales team performance. So you can actually see how each person on the sales team is doing. And over here you have the revenue per quarter. Now perhaps the most important part of this dashboard is that it's interactive. So anyone looking at this dashboard, can change and play with the data to get exactly the information they need. Up here you can change the date and the dashboard automatically updates to show you exactly the date you are asking for. You can also change the country or the region let say we just want to look at Latin America. Or the sales team let say, we just want to look at the enterprise sales team. Each time the dashboard automatically updates and tell you exactly the information you've asked for. This is a very powerful tool. >> As you can imagine now after seeing one, dashboard's have the potential to be very useful and compelling tools for companies. Let me tell you, especially when they are linked to databases, they can also save an analyst a lot of time and become an analyst's best friend. Let's hear what Tableau's dashboard functionality did for Ryan. >> So we used both Tableau desktop and Tableau server. So I think my favorite feature is that I can, at one point do an analysis, build a dashboard, publish it on the web, and allow people to interact with it whenever suits them, with data that refreshes on a reasonable time scale. So I can build something once, and make it available and refreshed all the time. So before I used Tableau, I had a job where I had to every Monday run a bunch of SQL queries and copy and paste the data into Excel and then run Pivot Tables on it, and then email it to someone. So I got back quite a lot of time by being able to say here is a dashboard that has all of this information, it's available whenever you want. And I also helped work on a tool to take a screenshot of that dashboard and email it to people so that they could get it in their inbox. >> You heard Ryan talk about Tableau desktop and server and me talk about the professional version of Tableau desktop. There are many different versions of Tableau, almost slightly different capabilities. There's a completely free version called Tableau Public which includes a Tableau Public interface you can have in your desktop. This version lets you do almost everything the other versions can do. But your data can't have more than 1,000,000 rows, you have to save your data to their website, not your desktop. And you cannot connect to databases, only to Excel spreadsheets, text files, and Microsoft access files. In my experience connecting to really big data files can also sometimes be a bit slow. There are also paid Tableau desktop versions, one for personal use and one for professional use. These both allow you to save data and workbooks to your desktop and ways to share your work with others via online tools. Luckily for us, Tableau has donated Tableau desktop subscriptions to our course, so you will get to experience first hand how to use them. In addition to saving files to your desktop with a normal subscription service, the personal version allows you to save data to Tableau online, Tableau server on the cloud. The professional version is the only version that can connect to databases, and therefore is really the only one that can efficiently handle those really big specialty datasets. It is also the only one that connect to Tableau server. Tableau server is basically a way to share things locally at your own company in a very secure and controllable way. No matter what version your company has available, Tableau is a very powerful tool that will dramatically improve the efficiency of your analysis. It will help you clean your data, find patterns in your data, test your hypothesis, and tell stories with your data. We've dedicated so much time to it in the specialization, because no matter whether you have a very strong quantitative background or no quantitative background at all, we truly believe Tableau will provide a lot of new value for you. With that, are you excited to become a Tableau expert? I hope so.
