Experimental


For humor (mostly), I have been known to join in the common refrain that no matter how creative and amazing and groundbreaking the dashboards are that you come up with, the customer will more often than not just want bar charts, line graphs, and crosstabs. ("We are so excited about Tableau! Now how can we use it to recreate these Excel charts we've been using for the last 15 years?")

Recently I was able to spend a few hours developing some unorthodox interactives, which used a similar type and shape of data to those used regularly by of one of our customers. Just for fun, I did some spit and polishing of my workbook, and brought it with me to a meeting with the client.

After showing and explaining the actual deliverables, I loaded up the unorthodox dashboard and said, "Now, for this one, we went a bit more ... experimental." (It involved bivariate choropleth tilemaps and heatmaps, but of course I can't share the actual data here. Just some unlabeled screengrabs.)


I proceeded to explain the concept behind that kind of visualization, the calculations I had applied to the dataset in order to create this kind of visual, and some of the insights that could be derived from it.

Honestly I expected it to be received tepidly at best. I wouldn't say that I'm fully jaded, but I will say that those of us who think about data visualization all day, every day, are more receptive to unusual visual encodings than the average person.

And, furthermore, I knew that if I were not there, verbally explaining the dashboard to the audience instead of dropping it in their laps without context, it would likely be too dense to grasp, even with loads of explainer text included.

However, I am pleased to say that in this case, not only were the analysts in the audience receptive to this interactive, but as it was explained to them in more detail, with certain features and top-level insights demonstrated to them in real-time, they became fully engaged and eager to investigate all of the data contained within it.

By the end of the meeting, it was clear that, while the client was satisfied with the closer-to-standard deliverables, it was the "experimental" dashboard that they were visibly excited about.


I think this is worth mentioning for a few reasons:

  1. Turning an "extra" dashboard around in limited time wouldn't have been possible without having had a lot of practice in developing creative products external to specific customer requests. Most of the functionality in the experimental workbook, as well as the entire concept of using this particular chart variant, originally grew out of Makeover Monday efforts. (Specifically, this one from a few weeks ago about Irish whisky.) PARTICIPATE IN MAKEOVER MONDAY, people. I can't think of a better communal experience that will give you the chance to try things without risk of failure, but WITH the potential to learn new design and technical skills.

  2. Don't be afraid to show some unusual, experimental, or challenging designs to your clients--but not to the *exclusion* of the expected (or required) deliverables. Sure, you have to meet the client's requests; but as a professional, it's part of your role to provide your expertise...which can and does include showing them additional solutions that may be preferable, albeit unusual. While they will sometimes (often) be set aside in confusion or distaste, they may spark other ideas about types of analyses, comparisons, and visualizations that were previously not considered, or thought possible. And in some cases, they might even be accepted as-is--and that's pretty satisfying.

  3. Try to show the complicated designs to people who really get deep into the details of the data. In the experience I had recently, the people who saw the value in the unorthodox charts were analysts--not executives. It shouldn't have surprised me that they were enthusiastic about being shown a new way to plumb their datasets...especially if having that capability could save them time, find them new insights, and differentiate their work from that of other analysts working on similar issues.

So, for those of you slogging through the data cleaning, crosstab-making, Excel-recreating muck, I say: hang in there. Keep offering up creative solutions whenever you can. Try to engage with the working analyst. And take whatever chance you get to experiment outside of work, so that when an opportunity arises, you'll not only recognize it, you'll know how to take advantage of it, technically and visually.

Thanks for tuning in to this edition of Life's Little Victories.

#tableau #bivariatechoropleth #heatmap #tilemap #experimental #analyst

0 views