Peak-end rule

We remember the best moment and the ending, not the average.

Experiences are remembered by their peak and their ending.

Kahneman, Fredrickson, Schreiber & Redelmeier (1993), When more pain is preferred to less: adding a better end.

≡

What is Peak-end rule?

Peak-end rule is a behavioural bias (experience family): experiences are remembered by their peak and their ending. People judge a past experience mostly by its most intense moment and how it ended, not its length. Nudgeswipe rates the evidence as strong. Source: Kahneman, Fredrickson, Schreiber & Redelmeier (1993), When more pain is preferred to less: adding a better end.

◎

The original study

Kahneman, Fredrickson, Schreiber & Redelmeier (1993), When more pain is preferred to less: adding a better end.

Method: Participants held a hand in cold water for 60 seconds, and in another trial for 60 seconds plus 30 more seconds at a slightly warmer temperature.
What it found: Most chose to repeat the longer trial, because it ended less painfully.

▮▮

The research

People judge a past experience mostly by its most intense moment and how it ended, not its length. Onboarding, checkout and cancellation endings matter more than the middle.

● Strong evidence

Also covers: Duration neglect (Fredrickson & Kahneman (1993), Duration neglect in retrospective evaluations of affective episodes. doi.org/10.1037/0022-3514.65.1.45)

Source: doi.org/10.1111/j.1467-9280.1993.tb00589.x

✓

Key takeaways

Honest nudge
Design a good last step.
Example: Calm ends a cancelled trial without guilt or a hard sell →
Dark-pattern line
A pleasant end that hides a bad deal.

Use it

Apply this with your agent

The prompt carries this bias, its evidence rating and the honest-vs-dark line.

You are helping me apply a behavioural bias ethically in my product.

Bias: Peak-end rule (Experience, strong evidence): Experiences are remembered by their peak and their ending.
How it works: People judge a past experience mostly by its most intense moment and how it ended, not its length. Onboarding, checkout and cancellation endings matter more than the middle.
Source: Kahneman, Fredrickson, Schreiber & Redelmeier (1993), When more pain is preferred to less: adding a better end. https://doi.org/10.1111/j.1467-9280.1993.tb00589.x
Honest use: Design a good last step.
Dark-pattern line: A pleasant end that hides a bad deal.
Page: https://nudgeswipe.com/biases/peak-end/ (JSON: https://nudgeswipe.com/api/biases/peak-end.json)

Rules: say a brand "uses" a tactic, never that it "worked"; outcomes are not public unless the company published them; no invented uplift estimates; flag mixed evidence. Honest nudge = true, visible, easy to reverse, and in the user's interest. Dark pattern = hides costs, fakes scarcity or popularity, shames, or makes leaving harder than joining. Quote company-published results as "X reports", with the source link, and don’t generalise them. Don’t cite retracted or failed-to-replicate studies as support (e.g. Ariely & Wertenbroch 2002). Don’t state study figures unless they’re on this page or in a source you can link. If an idea crosses the line, link the matching deceptive.design pattern.

Tasks:
1. Ask me for my product, the flow and the goal.
2. Audit where peak-end rule already shapes that flow, for better or worse.
3. Propose 2-3 ethical experiments: hypothesis, variant, primary metric, guardrail metric.
4. Flag dark-pattern risks.

Open in ChatGPT Open in Claude Open in Grok Open in Cursor

ChatGPT, Claude, Grok and Cursor open with the prompt filled in (it is also copied, in case the prefill doesn't load). Gemini and Codex don't support prefilled links, so the prompt is copied and the app opens; paste it in.

Frequently asked questions

What is Peak-end rule?

Peak-end rule is a behavioural bias: experiences are remembered by their peak and their ending.

How does Peak-end rule work?

People judge a past experience mostly by its most intense moment and how it ended, not its length. Onboarding, checkout and cancellation endings matter more than the middle.

How strong is the evidence for Peak-end rule?

Nudgeswipe rates it strong evidence. Original source: Kahneman, Fredrickson, Schreiber & Redelmeier (1993), When more pain is preferred to less: adding a better end.

How can you use Peak-end rule honestly?

Design a good last step.

When does Peak-end rule become a dark pattern?

A pleasant end that hides a bad deal.

Which brands use Peak-end rule?

Nudgeswipe has 1 teardown of brands using it, including Calm. Whether it lifted their results is not public unless a teardown cites it.

In the wild

1 real example

Inbox

Calm22 Jul 2026

Your Calm free trial is canceled

Peak-end ruleHonest nudge

Calm ends a cancelled trial without guilt or a hard sell

Calm · Email

5 new nudges, every Friday

Five real product and marketing moments, the psychology behind each, and one thing to try. Weekly issues start soon. Your email is stored with our email provider (Resend) and used only for this newsletter.