Peak-End Rule
The most intense moment and the ending can weigh heavily in how an experience is remembered.
People remember moments, not a perfect average of the whole journey.
A checkout works well until payment fails with a vague message. That moment becomes the emotional peak at the end. A precise explanation and a safe recovery path can change both the immediate outcome and the memory of the flow.
Origin
Fredrickson and Kahneman studied retrospective evaluation in 1993. Later work found strong support for peak and end effects, while also showing that the overall average experience can matter substantially.
Find the moment with the highest emotion or consequence. Improve the task there before adding celebration elsewhere.
Design the ending as a useful handoff. Confirm what happened, preserve the result, and make the next sensible action clear.
Recover carefully after failure. A specific explanation, retained input, and safe retry can soften a negative peak without pretending it did not happen.
Evaluate the whole journey. Peaks and endings influence memory, but they do not make repeated friction, exclusion, or delay acceptable.
Digital systems
A useful completion screen
After submission, the screen confirms the result, shows what happens next, and provides a receipt or reference. The ending closes the task instead of leaving a new question.
Physical signals
A clear hotel checkout
An accurate final bill, a simple key return, and a helpful farewell give the stay a clean ending. They cannot erase serious problems, but they prevent new ones at departure.
A failure or confusing final step carries more emotional weight than the routine moments that came before it.
Remove avoidable pain at the most intense point, then end with an accurate result, a clear next step, and a path to recovery when needed.
- Fredrickson, B. L., and Kahneman, D. (1993). Duration neglect in retrospective evaluations of affective episodes. Journal of Personality and Social Psychology, 65(1).
- Alaybek, B., et al. (2022). All's well that ends and peaks well? A meta-analysis of the peak-end rule and duration neglect. Organizational Behavior and Human Decision Processes, 170.