JH
Jonathan Haber

Add friction to unwanted behaviors

Inserting even a small obstacle between you and an unwanted behavior meaningfully reduces how often it happens.

Key takeaways

  • What it is: Inserting even a small obstacle between you and an unwanted behavior meaningfully reduces how often it happens.
  • Why it works: The same friction mechanism that helps desired behaviors also suppresses unwanted ones. Even a minor barrier — an additional click, moving an app off the home screen, keeping a food out of sight — introduces a pause where the automatic response can be interrupted. The pause is not willpower; it is a decision point that gives the prefrontal cortex time to re-engage before the habit fires.
  • Evidence: Backed by observational / correlational evidence.
  • Avoid: Adding friction that you quickly learn to navigate without thinking — friction works at first but habituates if it becomes a routine step rather than a genuine obstacle.

Why it works

The same friction mechanism that helps desired behaviors also suppresses unwanted ones. Even a minor barrier — an additional click, moving an app off the home screen, keeping a food out of sight — introduces a pause where the automatic response can be interrupted. The pause is not willpower; it is a decision point that gives the prefrontal cortex time to re-engage before the habit fires.

How to do it

  1. 1Identify the path of least resistance to the unwanted behavior and insert a step: unplug the TV and store the remote in another room; move junk food to the back of a high shelf.
  2. 2Log out of time-wasting apps and delete them from the home screen — even a 10-second sign-in adds enough friction.
  3. 3Do not rely on a single friction point; layer two or three small obstacles.

What the evidence says

Observational

Friction addition has direct support in food choice research (Wansink and colleagues) and digital behavior research: increasing the number of steps to access a behavior reliably reduces frequency. Note that some of Wansink’s food-environment research has been retracted; the core effect is replicated but specific effect-size claims from that line of work should be treated cautiously.

Honest caveat: Some friction research (especially Wansink lab) has faced reproducibility concerns. The effect is real and replicated by others, but exact magnitudes are uncertain.

References
  • — Cheung, Mittal & Bhattacharya (meta-analysis) on friction and food choice (see also Dayan & Bar-El 2011 for replications)

Common mistake

Adding friction that you quickly learn to navigate without thinking — friction works at first but habituates if it becomes a routine step rather than a genuine obstacle.

IX Coach helps you audit your digital environment and physical space to identify where unwanted behaviors have near-zero friction, then designs specific friction additions.

Practice this with IX Coach →

Related practices