The Definitive Guide to Tableau’s Analytics Pane (Part 1)

Constant Lines, Average Lines, Median with Quartiles, and Box Plots

In part one of our series on the Analytics pane in Tableau, Ethan does a technical deep dive on the Summarize section. Learn how to enhance your data visualizations with analytics by simply dragging and dropping them onto the view!

Hi, everyone. This Ethan Lang with Playfair Plus. And in today’s video, I’m going to be covering part one of a series about the Analytics pane. I’ll be covering the overview of the Analytics pane, as well as the summarized section of the Analytics pane.

So let’s jump over to Tableau. And I’ve built this supplementary workbook so you guys can follow along. You can actually download this from our Tableau Public, Playfair Data’s Tableau Public. But there’s a link to this exact workbook directly below this video. So feel free to click that, download the workbook, follow along.

You can also reengineer anything that I’ve done here. So if any of these functionalities of this workbook interests you, like I said, you can download it and re-engineer it yourself. But let’s take a look at the Analytics pane. And I’ll start by just opening up any sheet. So if we jump in, by default Tableau always lands us on the Data pane here.

And we can see on the Data pane, we’re all pretty familiar with this. We have our data sources here at the top. We have our measures and dimensions, our parameter listing. We have our sets, everything that we’re familiar within Tableau. Like I said, this is where we would land by default.

But at the top left of the authoring interface, you can actually toggle over to what’s called the Analytics pane within Tableau. By clicking this, that says Analytics. So if we click this tab, we’re now in the Analytics pane. So we can see that pane is now tabbed over to this pane.

And to kind of give us an overview of the Analytics pane, it breaks it into three sections here. So the first one is called Summarize. And this just has a very basic benchmarking tools that you can incorporate within your tool. You can add box plots or totals and subtotals within your workbook or onto the view.

The next section is called Model And this is where we start getting into some advanced analytics. They have different models that Tableau has incorporated within the tool that we can simply drag into our view and it will model out the data for us.

The last section is this custom section. And you can see here we have some very similar options to what we would find in the summarized section. We have things like reference line, reference bands, box plots. All of these we can find in the summarized section. But with the custom section, it gives us a little bit more flexibility with how we can customize those once we’ve kind of incorporated them within the view. And I’ll touch on that as we go through each one of these.

But to kick us off, I’ll start here at the top with Constant Line. If I drag Constant Line on to the view, we’ll be presented with these options here. And depending on where I drop this pill, where I drop Constant Line here, it’s going to add a constant line to a specific axis. So you can see here I have month of order date.

If I added that there, that’s in my columns. That’s going to add it to the x-axis. If I drop it onto SUM of sales, that’s in my row shelf. It’s going to add it to my y-axis. If I drop it on the table, though, it’s going to add a constant line onto both of those axes. So we can see both of those pills are now highlighted orange.

For this demonstration, I’m going to drop it on to sales. And I’ll simply just drag constant line and drop it onto that pill. And we notice here it’s added in a new constant line, which just appeared here at the bottom. And it defaults to the minimum value on the view.

Now, we’ll also notice in that same top left corner of the sheet, it’s now given me the option to type in what the value will be for that constant line. So essentially, we’re adding a constant value to our view, so we can type in a constant value here.

If I just match that, I’ll put in 50,000 to just demonstrate the differences here. So now I have a constant line of 70,000, which I had already previously added to the view. And now I’ve just added a new constant line of 50,000. So now we can see both of those in action here. And if I hover over anywhere on the sheet, you can see what that value is.

One more thing to cover on constant line. Not only is this a great way to set benchmarks, so I can kind of see what’s above and below that constant line, I can also use that same value to start building logic, where I could color or add some aesthetics that would kind of conditionally format if I’m above or below that reference line.

Another great tip if you right-click anywhere on the constant line, you’ll be presented with these options. You can remove that constant line. You can edit it and format it. You can move it to the back or to the front. So if you had, say, this line chart, if I wanted to move it behind the line chart, I can do that.

If I wanted to move it in front of the line chart, I can select that other option. If we edit the constant line, we’ll be presented with this set of options here. And this allows us to change the value that we set. We can change where it is. So we can set it to a minimum or maximum.

That will automatically default that value, depending on what the min and max are. Average median. We can get a sum of some sort of or one of our measures. We can also do a total, where it adds in a total, running total of what’s in the view. So we can see that in action here.

Now, I’ll just leave it to constant. That’s what we’re covering right now. You can also change the label. So notice here, I have 70,000 on the label. I can remove that label, make it none. So we can see that 70,000 disappear. I can add in a computation or add in a custom label.

So let’s say I wanted to say the value, but something like benchmark. I can type in benchmark there and then add in the value by clicking on this little caret here. And I can select value from that menu. So now I have benchmark 70,000 as my custom label.

We can also format the lines. So if we want to make that pop out a little bit more, we can change this to more of a red by selecting our color options here. We can change it to, let’s say, a dotted line versus a solid line. We can also make it a little bit more opaque. And I’ll select OK here and we’ll see that take effect.

Now, our last thing, we can format. If I select Format when right-clicking here, we’ll notice the Analytics pane is now replaced with our format reference line. These give us the same options that we saw earlier, but slightly more options. We can now change the font. We can add in alignment.

We can change the format of the number itself. All of those options can be displayed from right here by selecting Format from the menu. That’s constant line. Now, we can also add in a summarize option. It’s very similar to the constant line. I’m actually going to clear these from the view so we can see what this looks like.

If I drag an average line onto the view, notice that we’re presented with something slightly different than constant lines. So here, when it says add a reference line, I now have the option to add it to the entire table. And that means it’s going to calculate that average using every value in the table.

If I had separate panes, I could calculate the average across each pane, or I can add in an average across each cell, if I had separate cells. For this demonstration, I’m going to calculate it over the total table. So I’ll just drop average line onto table. And it’s going to add that right here. So we can see the average of all the values in my view currently is 47,858.

Again, if we right-click on that average line and click Edit, we can change some of those options here. This first is the scope. So again, we can calculate it over the entire table, which is where I dropped it. We can calculate it over the pane or per cell. I can also change the value or what’s being calculated as the value.

So currently, I only have the sum of sales in my view. So that’s the value that I can set. But I can also choose a parameter, a bin, or a set. Again, if I click on this dropdown, I can see the other options here. I’ll leave it as average for this demonstration. We can change the label, just like we did with the constant line, but I can also change the tooltip.

And then, here, I can either select the line only. Or I can start changing it to more of a model, which we’ll touch on later, by adding the line and the confidence interval. And then the last option here I can add the interval only, which will draw a band.

Lastly the formatting. We can format the line itself. We can also fill above or below the line. So to demonstrate what that would look like, if we fill– and I’ll just choose this blue– above the line, it’s going to fill everything above the line and it’ll draw this distribution band.

Or below the line, same thing. Select none. And then I’ll click OK. So the great thing about the constant line and average line that I love is it’s a great way to add a benchmark into your view, especially, again, if you use them in tandem. So we can add the constant line to the view. We can add an average line to the view. And we can see how we’re benchmarking against those values in our view.

As our average moves up, we can work our way towards the benchmark. Or if we’re above it, we can make sure we’re staying above it and not dropping below that benchmark. So it really brings in this very unique analysis within our view that we can do directly from just dropping in these lines as reference lines onto the view.

Now I’m going to clear average line and I’m going to drop in the median with quartiles. So when I drag this onto the view, again, we’ll be presented with the same options as we were with the average line. We can calculate the median across the table, the pane, or individual cells. For this demonstration, I’ll drop it onto table.

And we can see it’s going to add that median line, but also with these quartiles. So we can see the upper quartile and the lower quartile of the values within our view. Again, if I right-click, I’ll be presented with those same options as I were with constant and average line. I can edit. I can format. And all of those options are the same as before.

Now, one thing I like to do, especially if I’m getting to know the data, I’ll drop in the median, but I’ll also drop in an average line. And what this will show me is if these two lines are pretty close to one another, which we can see they fairly are here, our data is pretty normally distributed. It kind of follows a nice pattern, if you will.

However, if my average and median line were pretty far apart, there might be some very heavy outliers in my data that I might need to take a closer look at. For instance, we can see that this data point up here, it might be leading to some of those outliers. That’s going to be pulling our average line farther up away from our median, which we can kind of see that here.

But with it falling below that upper quartile still, to me I don’t think that represents much of an issue. Maybe something we don’t necessarily dive into too in depth. OK, I’ll clear those. And we’ll move on to box plot.

And for this, I’m actually going to switch over to box plot here. So I’m going to hop back out into my main view. I’m going to select the box plot from my Analytics pane selection here. And then I’m going to jump into this sheet.

So now that I’ve selected box plot in my view, we can see that this is now grayed out. That’s because it’s already created a box plot form here. However, if this was a fresh view, I could simply drag in box plot, and it would create this data visualization for me. It’d be the same thing as selecting it from the Show Me tab.

So if we click the Show Me box-and-whisker plot, it would build out the same view. However, the Analytics pane actually gives us slightly more options here. So if I right-click on any of the box-and-whisker plots, I can then edit. And I’m presented with this menu here.

So we can see that it’s created a box plot. The box plot options I can change. So data within 1 and 1/2 times or maximum extent of the data. I can also select this option, which is going to hide all of the values within the boxes itself. So if I select that and then click OK, we’ll see now here that I only have this one outlier that falls outside of the range of my box-and-whisker.

We deselect that option. So we can see those come back. So I’ll deselect that. And we can see now all of those– or all of those data points within the view have now come back. So we can see them here. And that’s a great way, again, to just find outliers within your view.

So if I select that again, I can see now I only have that one outlier. Now I can start investigating that data point a little more. Maybe I can view the data, view all the underlying data to figure out what’s causing that.

So for now, I’ll just quickly go back into the Edit options, and I’ll uncheck that box. That’ll bring in all the data points back into the view. And then I’m going to cover the formatting here. So we can see I have these different styles I can choose between.

It defaults to Modern, but I can toggle to Glass, and you can see it just updates the format kind of holistically by just selecting these different styles.

I can also choose or change the formatting of the fill. So right now, it’s a light gray. I can change that to green or orange. Whatever I wanted to represent here. Maybe you have certain branding that you want to incorporate. I will say that the options are rather light.

So just use your best judgment on the colors that you’re using for the dots and what makes the most sense to the workbook that you’re working in. I can change the borders. So just like changing a line, I can change the borders here. And then the whisker. I can also edit or format the way the whisker looks.

So if I want to make it a little bit smaller, wider, no whisker at the end of the line. I can choose all of those options directly from this formatting tab. For now I’ll just click OK, and that’s our box plot.

The next one I’ll cover is Totals. And again, I’m going to hop back over to the workbook and toggle over to Totals. And Totals, notice that it was grayed out throughout this entire demonstration. But here, it’s actually solid. Primarily, you want to add Totals if you’re working with some sort of cross tab view.

So we can see here I have this cross tab view. It’s looking at Region by Sub-Category and then showing me the SUM of Sales for each Sub-Category. Now, I wanted to include a Totals at the bottom of these. So you can see each Sub-Category currently has a Total section here. If I grab Total from the Summarize, you’ll see we’re presented with some of those similar options that we had saw before.

So I can aggregate the Totals by a subtotal, or the column grand totals, or row grand totals. And this would depend on how that cross tab is formatted and how it’s built. So notice right now, I cannot drop it on the subtotals. I cannot drop it on the grand totals. But I can drop it on the column grand totals.

All that is is that’s because the way that this cross tab is built. If I switch to this view, and I did that by just clicking on here where it’s going to swap my rows and columns, now if I drag subtotals in, notice my Column Grand Totals is now shaded gray but my Row Grand Totals are an option.

Again, it’s all dependent on how the view is built and where it can add in those totals. So I’ll swap it back to Rows. And that covers Totals. There is another way that you can implement this outside of the Analytics pane, which I think is what people are most used to. If you select the Analysis tab at the top, you can add Totals from there and you’re presented with those options.

This has been Ethan Lang with Playfair Plus. Thanks for catching part one of our series on the Analytics pane, and be sure to catch part two of our series. Thank you again for watching!