False friend: beware disfluency as a BE tool

Disfluency, in a behavioral economics, means making a message harder to read with the intent that its readers will either:

  • remember it better
  • perform the task described in the message better

I’m taking the opportunity of an excellent post by Andrew Gelman to highlight this false friend as a tool you should probably not use, or use with extreme caution.

You can go read Andrew’s post for the details, but basically, it’s an example of how flaky some studies really are, and at the latest count, there has been 19 studies done and only 2 find a statistically significant effect, and on very small sample sizes at that.

So in spite of what some popular behavioral economists are telling their book readers to do, you should most likely avoid making your fonts harder to read if you’re running Conversion rate Optimization programs for instance. Nothing in terms of scientific results support that, no matter what they say.