Avoid negative social proof (the "everyone is doing it wrong" mistake)
Describing a problem’s prevalence can inadvertently normalize it.
Key takeaways
- What it is: Describing a problem’s prevalence can inadvertently normalize it.
- Why it works: When a message highlights how common a bad behavior is — "35% of people fail to recycle," "most people skip their check-ups" — it unintentionally communicates that the bad behavior is the norm. The descriptive norm (what people do) can override the injunctive norm (what people should do), which is why highlighting problems by their frequency can increase rather than decrease them.
- Evidence: Backed by randomized trials / meta-analyses.
- Avoid: Campaigns that lead with how widespread a problem is ("most people are X") when the goal is to reduce X — a well-documented pattern in public health messaging that can increase the very behavior it targets.
Why it works
When a message highlights how common a bad behavior is — "35% of people fail to recycle," "most people skip their check-ups" — it unintentionally communicates that the bad behavior is the norm. The descriptive norm (what people do) can override the injunctive norm (what people should do), which is why highlighting problems by their frequency can increase rather than decrease them.
How to do it
- 1Before any "the problem is widespread" framing, check whether it implies the behavior is normal.
- 2Pivot the statistic: instead of "35% fail to recycle," say "65% of people in our community do recycle."
- 3Pair a negative statistic with a clear injunctive norm: "though 35% miss it, the standard is X."
What the evidence says
RCT / meta-analysisSchultz et al. (2007) ran a field experiment showing that households above average energy use reduced after receiving peer comparison, but below-average users increased (the boomerang effect) — corrected by adding an injunctive norm signal (a smiley face). This directly demonstrates the negative-social-proof hazard.
Honest caveat: The energy-use context is specific; the generalization to other behavior domains is well-reasoned from social norm theory but should be tested domain by domain.
- — Schultz et al. (2007), The constructive, destructive, and reconstructive power of social norms, Psychological Science
Common mistake
Campaigns that lead with how widespread a problem is ("most people are X") when the goal is to reduce X — a well-documented pattern in public health messaging that can increase the very behavior it targets.
IX Coach frames your patterns against what people who are succeeding with this challenge do — not against the failure rate — so the reference point pulls you toward the goal.
Practice this with IX Coach →