Learn how to perform a basic exploratory analysis in Tableau.
- [Instructor] I think of Tableau in a similar vein as Photoshop. They're two totally different programs that do two totally different things, but what I mean is that Photoshop is a critical tool for certain aspects of design and photography. Tableau is like that for data analysis. It's a go-to kind of tool. There's a lot it won't do, and that's why we have included the other tools in this course. But for what it does do, it does a pretty awesome job at it. I already have Tableau open and ready to go here and we're going to connect to our data so I'm going to click on Text File to connect.
I am then going to navigate to our Exercise Files and I'm going to click on exploratory-tableau.csv. Click on Open, that takes a second, and it's going to bring our data into the Tableau environment. Now that we have our data in, we can take a quick look at what we have. We can see that we have Keywords, we have number of Clicks, we have Click-through Rate, Conversion Rate, Cost Per Action, and the number of Impressions for each of those keywords.
So that looks good, and now let's click on Sheet 1 where we can begin to visualize this data. So one of the first things I want to do here is I want to grab the Keywords dimension and I'm going to drop that onto our Rows, so now we can see our Keywords here. Now, I want to grab the Impressions from our Measures and just drop that onto our Color Marks tab. So pretty much instantaneously, we can see a heat map for this data so ultimately this is showing us the impressions for each of these keywords.
So if I just mouse over here, I can see for Low Price Air, 354,000 impressions roughly. And then a lighter color, less impression. So Ticket Best Price, roughly 274,000. So again, it's a great tool for doing this kind of exploratory analysis really quickly. And in this particular instance, we were looking at the Impressions, but we could quickly take a look at CPA. And I'll just replace that. So, all I did was grab that particular measure, dropped it on the Color Marks card, and it has now replaced that analysis there.
So the colors that we see on this particular visualization are nice, but maybe one of the questions that we want to ask is how meaningful are they? So in this particular instance, we are looking at the Cost Per Action and it's advantageous for us and for our campaigns and for the return on investment for our campaigns, first to get the Cost Per Action or cost for acquisition as low as possible, and so here we can see that the darker blue color shows us a Cost Per Action of roughly $27 and the lighter color in this range shows us a Cost Per Action of $5.67.
So what this is saying is that the darker the visualization, the darker the color on this particular heat map, the more it's going to cost. Well, Tableau gives us a lot of tools that we can leverage to use to help tell a story with our data. And so, I'm going to come over here and I'm going to actually collapse this Show Me bar and I'm going to look at our overall color palette here. And I'm going to do this dropdown and I'm going to select Edit Colors. And we can see when I drop down the Palette option, that there are a number of color palettes pre-defined in Tableau that we can use.
And so something like a CPA, we might say, well it really works on a range of something that's good and something that's less than good. And so something that we might look at is a red, green, gold, diverging color palette to help make some meaning of this. So in this case, what this is showing is that on the lower end of that range, the visualization will be red, and on the higher end of that range, it'll be green. And really, the inverse is true here for Cost Per Action. If we were looking at Click-through Rate or Impressions, this would be a good continuum.
But, for this particular Cost Per Action, we're going to reverse that around. So we select Reversed and then Okay. And so this completely changes the story that's being told here, we can see very quickly that Low Price Tickets Airline is a keyword that we're paying quite a bit for and it's really outside of the range of some of these other keywords. Let's take a moment and take this visualization from the top, if you will. So I'm going to remove Keywords.
I'm just going to select it and hit Delete. I'm then going to remove the CPA as well, select it and hit Delete. Now, wouldn't it be nice to do a data visualization similar to what we produced in R, so we can see all of our data really side by side? Well, we can do that in Tableau, quite easily, really. I'm first going to select our Keyword dimension, I'm going to drop that on our Rows, and at this point, we want to convert our campaign data from a Measure to a Dimension.
So what we can do here is select the dropdown, we can Convert to Dimension and just that easily, that particular component becomes a Dimension as opposed to a Measure. So I'm going to take Campaign, I'm going to put that on the Columns. And now we can see each of our campaigns on this grid. I'm now going to select Cost Per Action and drop it on our Colors Mark card. And now instantaneously we can see across each of our different campaigns how well our CPA is performing.
So again, this gives us a visualization that we can step back from, we can analyze, we can assess and we can identify opportunities for creating optimizations within our campaign. Let me be the first to congratulate you with the completion of this video, you have now earned your white belt in Data Science for Marketing. Next up, we're going to take a step back and assess the pros and the cons of these platforms.
In this course, discover how to gain valuable insights from large data sets using specific languages and tools. Follow Chris DallaVilla as he walks through how to use R, Python, and Tableau to perform data modeling and assess performance. As Chris dives into these concepts, he shares specific case studies that come directly from his own work with clients. Plus, he shares three essential—and practical—best practices for data-driven marketing that you can use to bolster your organization's marketing performance.
- Installing R, Python, and Tableau
- Navigating the UI for R, Python, and Tableau
- Using R, Python, and Tableau
- Exploratory analysis
- Performing regression analysis
- Performing a cluster analysis
- Performing a conjoint assessment
- Stakeholder alignment