In Chapter 5 of Cole Nussbaumer’s new book, Storytelling with Data, she shows a revised version of a stacked column chart that highlights three segments with labels off to the left and two summary numbers to the right. It looks something like this:
Cole writes, “Being smart with color, aligning objects, and leveraging white space brings a sense of visual organization to your design. This attention to aesthetics shows a general respect for your work and your audience.” I couldn’t agree more.
So, I thought I might take an opportunity to show you how I would make this graph in Excel. If you follow the Excel tutorials on this blog, you’ll know that I always try to use data to add labels, markers, and other chart elements. I try to avoid adding text boxes or drawing lines when building charts in Excel because I find they rarely end up exactly where I want them, it’s harder to move them over to PowerPoint, and, most importantly, I will have to manually move things around when I update the data. (Note: I’m using the Mac version of Excel 2011 for this tutorial. It works the same way on the Excel 2010 Windows version, but will be a bit different on the 2013 or 2016 versions of Excel.)
The basic data include two values for seven different Segments. The default chart with data labels on each slide looks something like this:
What Cole does is to highlight just the three segments that are of particular interest. She also adds labels to the left of the left-hand column. Importantly—and what I think is particularly great about her redesign—is the “30%” and “50%” labels on the right of each column. Those labels aggregate the three slices so the reader can see both the parts and the whole of each column.
One way to implement the Segment labels is to add another series that sits to the left of the original two. I then label each slice with the Segment name and set the Fill of each section to No Fill. The primary problem with this approach is that the labels are too separated from the rest of the chart and the horizontal axis extends all the way to the left, which is not the case in Cole’s remake.
So how did I make Cole’s chart in Excel? I added eight data points to the chart and changed each of them to a scatterplot.
The first pair of points (under the “Number Scatters” header) are used to add the “30%” and “50%” labels. I add the vertical lines next to these labels by using vertical error bars (using the “PosError” and “NegError” data values). The x-values for these two series (1.22 and 2.22) were chosen so the points sit just off to the right of each column. The y-values sit right in the middle of the three highlighted series (and constructed via a formula in the cell). The two have the value names (30% and 50%) and are entered in those cells via formula.
For the seven “Segment” labels to the left of the chart (under the “Label Scatters” header), the x-values are set to sit in the position to the left of the chart, and the y-values are equal to the Segments values.
You can see the scatterplot points in this version. I add the data points and then the labels, positioning them to the right or left of the point, as desired. I then go back and set the Marker Style to No Marker to hide the marker leaving just the label.
So that’s it, really. Nothing too spectacular. I just used scatterplots to encode the data on the chart.
Do you have an alternative way to make this chart? Please let me know on Twitter or in the comment box below.
Oh, and be sure to listen to my interview with Cole about her new book on Episode #24 of The PolicyViz Podcast.