Graph from the "don't try this at home" drawer or a story about how the user sometimes doesn't know what he needs.
Several years ago, I was asked to add a graph to the sales dashboard. Its purpose was to show to the salesmen the relationship between sales performance and customer profitability.
A manager who characterized the graph was very experienced. It seemed he knew exactly what he needed. He even drew the sketch for me, and I thought it's really cool. The result – 2 graphs as shown below.
Most of the sales team members who used this dashboard only exported data from it into Excel.
A few of them tried to gain insights from it and turned to me with clarifying questions.
What went wrong with these beautiful graphs?
1. A single place with contradictory messages can be confusing.
As an example,
A pair of bars of the same length, but in different colors - one green (good), the other orange (bad).
2. There is too much "noise".
Dashboard's purpose is to show issues to improve, so these are the areas that need to be highlighted. In this one there are colors of traffic light. Every single one of them demands attention on its own. However, when all three appear together, a decoding operation is required to understand the differences.
Another challenge I added was to separate shades in addition to color differences.
3. The most important statistic of profitability that needs to be rated is not shown - users need to encode it by the color.
As a result, there is no chance to understand the relationship between profitability and sales, which was the purpose of the graph...
Which alternatives are available?
Create a visualization that allows you to:
Analyze all indicators together.
Put them in the correct order of importance.
Using color effectively for emphasis and warnings.
Provide all the necessary data without guessing.
Comentarios