Visualizing Data For Impact

We work extremely hard on collecting data. Then more time reporting the slices. Finally, even more time on doing real analysis to come up with powerful insights.

And then, it dies at the last mile.

The most common reasons are that we've not really thought through clearly what action we are trying to drive, which makes for an unnecessarily complicated story, which ends up being manifested with a visual that is difficult to understand.

Here's a representative example, as you reflect on it try to identify the insight being communicated...

Did you understand what the slide is saying? If you were the leader, would you know where to dig deeper or what action to take?

While this visual is the result of a lot of hard work, it is complex and difficult to understand. It is also puts the Analyst/Marketer who's presenting it in the tough position of having to over-explain while trying keep attention of the leaders.

With some focused effort, we can close the last-mile gap.

First, we can only show the data that is required. The goal in the original was to show top priorities for 12 months, if so showing the data for Aug 2017 adds extraneous burden to the graphic.

Second, we can experiment with alternative visualization options (easily available in Excel or d3js). In this case, we have five dimensions of data with two data points each. If all we want is a comparison of two data points for each dimension, we can possibly use the radar chart.

Applying those fixes, here's the end result...

It is easier to see the two data points, and realize that only two are expected to change (sadly trusting relationship is going down!).

Since the data is simply presented, the time available in the meeting can be used to for a discussion of what actions the company should take.

Simplicity powers data's influence.

In a recent blog post, Closing Data's Last-Mile Gap: Visualizing For Impact!, I'd used using 17 examples, like above, to illustrate 11 principles you can apply to close the last-mile gap:

1. Graphing choices can exaggerate or undersell reality. 
2. False precision can cause clutter, and undercut the Analyst’s brilliance.
3. Remove the distractions, ruthlessly.
4. Show as much data as is required, and no more.
5. Experiment with visualization options, even in Excel!
6. Don’t send a graphic to do a table’s job.
7. Please, please, please keep the end-user in mind.
8. Eat Pies, Don't Share Them!
9. Make your tables pop, guide the reader’s eye.
10. Let the higher order bit be your north star.
11. Why. Your job is to answer why!

The last-mile gap is a key contributor to the lack of impact of your data efforts. Closing that gap requires the application of above principles, but it also requires time to be set aside to invest in thinking critically about the story.

The rewards are: True competitive advantage for your company. Faster promotion for you. :)

If you would like to assess how good your skills are at closing the last-mile gap, visit my blog post to tackle an example even I could not fix!

Good luck.