Positive Deviance Framework
Positive Deviance finds the people already getting better results than their peers, inside the same constraints, and spreads what they actually do, instead of importing a solution from outside that never quite fits.
The four D's run in sequence, from defining the problem through to spreading what the outliers already do across the whole group.
Reach for this when…
- An outside best-practice keeps failing to take root because it does not fit local conditions.
- You suspect someone in the organisation has already solved the problem quietly.
- A behaviour-change programme keeps stalling because people resist being told what to do.
How to run it
- Define the problem and the outcome you actually want, in the community's own terms.
- Determine what common practice looks like right now.
- Discover the positive deviants who are getting better results with the same resources, and the specific, unusual behaviours behind it.
- Design ways for others to practice those behaviours themselves, and let results spread by peer example.
A worked example
Situation. Sunita Patra ran a health clinic network across rural Koraput district, Odisha, India, where a standard nutrition programme had failed to shift child malnutrition rates for two years.
Applied. Instead of another outside programme, her team found a handful of mothers in the same villages whose children were healthy despite the same poverty, and studied exactly what those mothers fed and did differently.
Result. Other mothers were invited to cook alongside the positive deviants rather than attend a lecture, and malnutrition rates in the pilot villages fell within six months.
The catch
Positive Deviance only works where deviants genuinely exist within the same constraints. If nobody is actually beating the odds, there is nothing to find and spread. It is also slower than importing a ready-made programme, and it depends on real trust to get people to show you what they actually do, not what they think you want to hear.
If your 'positive deviant' has resources the rest of the group does not, you have found an outlier, not a transferable behaviour.
Origin: Jerry Sternin and Monique Sternin; concept from Marian Zeitlin