Menu

A Data Visualization Opportunity

cover-image

Critical review of the data visualization practices at Quantcast.

Quantcast offers "insights into how web properties are faring." Rather, I would say, a perplexing mix of poorly chosen chart types, a cryptic scoring system, and data visualization techniques that require too much effort to understand. Stephen Few would be displeased.

Let's identify the opportunities. In future posts, I'll come back to remake these visualizations.

Traditional Chart Types

The Stacked Area Graph is widely criticized for its shortcomings with part-to-whole comparisons. In the Quantcast area chart, what to do if "Change in Mobile" is the pattern I'm interested in isolating? I'm given no tools to do so visually. Online traffic drives the entire visual, simply by being at the bottom.

Dan Murray has pointed me to a new tool called Visage, which addresses this question with an elegant use of transparency. It would be great if Tableau were able to incorporate these techniques, as well!


"Stacked area charts work when you primarily want to show how the whole changes through time and only give a rough sense of how the parts compare.." Stephen Few


Then come the Pie Charts. For a detailed review of why Stacked Area Graphs and Pie Charts fail at part-to-whole comparisons, see this article from Stephen Few, 2011.

Coloring the tips on the demographic bars is a nice feature. Pie charts, not so much.

While coloring the tips of the bars that extend beyond the reference line is a nice visual feature that helps to quickly identify the demographics that are over-indexing, what's perplexing is the cryptic INDEX system itself. The result is a visual that looks flashy but is difficult to interpret.

The index represents "the delivery of a specific audience segment compared to the internet average of 100". As an example: If a property indexes 100 for age 18-24, that means a given visitor is as likely to be 18-24 as any internet user chosen at random.

An index of 200 means the visitor is twice as likely to be 18-24, 50 means half as likely, and so on.

By this reasoning, it follows that "an INDEX of 84 means a given visitor to the website is 16% less likely to be of a certain demographic than any internet user chosen at random."

This is confusing. In the real world 84 -16 .

Tool-Tips and Legends

The screenshot below was chosen from the Quantcast data for the web property Zimbio.com. Hovering over the question mark for help indicates there are three types of visitors: Addicts, Regulars, and Passers-by.

Without some serious study, these charts don't communicate much at all. The bars have neither a color legend nor a tool-tip. And just what does each bar represent?

As an exercise: set yourself a timer & study this chart above. How long does it take to interpret the data?

Use of Color

Finally, the last set of charts make two comparisons each, for two geogrophies. Problems with the visual layout aside (which we'll address in a future make-over), at this point I only want to focus on the use of color.

Stop the Bleeding

Color (Hue) is excellent for associating quantitative values to categories. Similar to the use of blue in the bars above, that efficacy is lost here. The single red hue is used in all four of the bars.

What's more, in Western cultures, a red hue is associated with one or more of the following: Stop, Alert, Bad, Problem, or Danger. According to Amy Morin, in How To Use Color Psychology To Give Your Business An Edge, red is associated with Rage, Anger, and Annoyance.

While those emotions may be quite valid after trying hard to interpret the visual display of information at Quantcast, in the context of this graph none of the data has any reason to be associated with a bright red hue.

In Summary

The data vis techniques at Quantcast have opportunities for improvement. Over the coming weeks, I will give these data visualizations a make-over.

Word Count: 665

References

  1. "Reading Our Reports", Quantcast.com: https://www.quantcast.com/help/how-to-read-our-reports
  2. "I hate stacked area charts", Dr. Drang, And now it's all this, November 22nd, 2011: http://www.leancrew.com/all-this/2011/11/i-hate-stacked-area-charts
  3. "A Tool For Building Beautiful Data Visualizations", Margaret Rhodes, Fast Company, May 12, 2014: http://www.fastcodesign.com/3030419/from-the-designers-of-mintcom-a-tool-for-building-beautiful-data-visualizations
  4. "The Worst Chart In The World", WALTER HICKEY, Business Insider, June 17, 2013: http://www.businessinsider.com/pie-charts-are-the-worst-2013-6
  5. "Reading Our Reports", Quantcast.com: https://www.quantcast.com/help/how-to-read-our-reports
  6. "Color the Tips!", Keith Helfrich, Red Headed Step Data, May 14, 2014: http://redheadedstepdata.io/color-the-tips/
  7. "Data for Zimbio.com", Quantcast: https://www.quantcast.com/zimbio.com
  8. "How To Use Color Psychology To Give Your Business An Edge", Amy Morin, Forbes, February 4, 2014: http://www.forbes.com/sites/amymorin/2014/02/04/how-to-use-color-psychology-to-give-your-business-an-edge