The Peak-End Rule: Why Customers Remember Two Moments

Most customer experience programmes are built on an averaging fallacy. They measure every touchpoint, calculate a mean score, and optimise the middle. The problem is that human memory does not work that way. Customers do not remember the average of an experience — they remember two moments: the emotional peak and the ending. Everything else, however carefully designed, fades.

This is the peak-end rule, and it is one of the most consequential findings in behavioural science for anyone responsible for designing or managing customer journeys.

What the peak-end rule actually says

The peak-end rule is a cognitive heuristic, first described by Daniel Kahneman and colleagues in a 1993 paper published in the Journal of Experimental Psychology: General, titled "When More Pain Is Preferred to Less: Adding a Better End." The research demonstrated that people's retrospective evaluations of an experience are disproportionately influenced by two moments: the most intense emotional point (positive or negative) and how the experience ended. Duration — how long the experience actually lasted — had surprisingly little effect on recalled satisfaction. Kahneman labelled this the "duration neglect" effect.

The original experiments used cold-water immersion tasks. Participants who endured a longer trial — but one that ended with slightly warmer water — rated the overall experience as less unpleasant than a shorter trial that ended at peak discomfort. Objectively worse by total suffering. Subjectively better by remembered experience. The implication for service design is not subtle.

"Customers do not evaluate what happened to them — they evaluate what they remember happening to them. Those are not the same thing, and the gap between them is where most CX programmes quietly fail."

Why this matters more than your average NPS score

The standard CX measurement stack — NPS, CSAT, CES — is typically collected at single touchpoints or at the end of a defined interaction. These tools capture a signal, but they do not tell you which moments are constructing the memory that drives future behaviour. A customer who scores a 7 on a post-service survey may have experienced a genuinely excellent resolution moment buried inside an otherwise mediocre journey. That resolution is doing the heavy lifting on loyalty — but your dashboard does not know it.

Conversely, a customer who rates an experience 8 out of 10 overall may have encountered a sharp negative peak mid-journey — a rude handoff, a confusing form, a broken promise — that your aggregate score has smoothed over. That peak is quietly eroding the relationship. The memory of it will surface the next time a competitor makes a credible offer.

The peak-end rule reframes the measurement question. Instead of asking "how did we score across the journey?", the more useful question is: "what is the emotional peak we are creating, and how are we ending?" If you cannot answer both with specificity, your customer journey design is operating without its most important inputs.

How the peak-end rule plays out in real service contexts

Consider a hospital discharge process. A patient may have received attentive, competent care across several days. But if the discharge experience involves an hour of waiting in a corridor, confusing paperwork, and a hurried goodbye from a nurse who is clearly occupied elsewhere, that ending will colour the entire memory of the stay. The clinical quality — the genuine peak of care — is undermined by a weak ending that costs almost nothing to fix.

The same mechanism operates in banking and financial services. A mortgage application that ends with a clear, personalised welcome message, a named relationship manager, and a small but thoughtful gesture — a handwritten note, a call from a senior adviser — will be remembered more warmly than an identical process that ends with an automated email and a PDF. The functional outcome is the same. The memory is not.

Retail offers an instructive counterexample. The queuing experience at checkout is a well-documented source of negative peaks. Retailers who have invested in queue management — not to eliminate waiting, but to make the final moments of the wait feel shorter and the checkout itself feel warmer — have shifted the ending experience without touching the rest of the journey. That is the peak-end rule applied as a design constraint, not a philosophical observation.

Designing for peaks: deliberate, not accidental

Most organisations create emotional peaks by accident. A complaint that gets resolved brilliantly becomes a positive peak — not because the company planned it, but because one employee made an exceptional call. A billing error that goes unacknowledged becomes a negative peak that outlasts a dozen pleasant interactions. Neither outcome was designed. Both are shaping customer memory and future loyalty.

The more disciplined approach is to treat peak moments as design objects. This means identifying, in advance, which moments in a journey carry the highest emotional charge — either because they involve high stakes (a first purchase, a complaint, a renewal decision) or because they are inherently emotionally loaded (a welcome, a farewell, a moment of vulnerability). Then engineering those moments with the same rigour applied to a product feature.

In behavioural economics terms, this is choice architecture applied to emotional experience. You are not leaving the peak to chance — you are structuring the environment so that the most memorable moments are the ones you have designed to be memorable. The tools available include:

• Elevation: adding unexpected positive elements at a high-stakes moment — a surprise upgrade, a personalised acknowledgement, a gesture that signals the customer has been seen as an individual.
• Pride: creating moments where the customer feels recognised for a decision or achievement — a loyalty milestone celebrated meaningfully, not with a generic email.
• Insight: giving customers a moment of genuine clarity — a summary of what they have achieved, a personalised recommendation that shows you understand their situation.
• Connection: a human moment that breaks the transactional frame — a staff member who remembers a preference, a message that references a previous interaction with specificity.

These are not expensive interventions. They are attentional ones. The cost is in the design discipline required to identify the right moments and train the people or configure the systems that deliver them.

Designing for endings: the most under-invested moment in CX

If peaks are under-designed, endings are almost entirely neglected. Most journeys end with a confirmation email, a receipt, or silence. None of these is a designed ending — they are administrative closures that leave the emotional arc of the experience unresolved.

A designed ending does three things. It signals completion clearly, so the customer's mind can close the loop on the experience. It reinforces the positive peak — reminding the customer of the best moment, not the most recent one. And it creates a bridge to the next interaction, so the relationship continues rather than resets.

In hospitality, the best operators understand this instinctively. The checkout experience at a well-run hotel is not an administrative formality — it is a deliberate emotional landing. The staff member who says "we noticed you enjoyed the terrace — we have saved your preference for your next visit" is not following a script about upselling. They are closing the experience on a note of personal recognition that will be disproportionately influential in how the guest remembers the entire stay.

The same principle applies to digital journeys. An e-commerce confirmation page that simply lists order details is a missed ending. One that acknowledges the purchase with warmth, surfaces a relevant piece of content, or offers a small, unexpected benefit creates a different emotional close — and a different memory.

The duration neglect trap in CX strategy

Duration neglect — the finding that how long an experience lasts has little bearing on how it is remembered — has a direct implication for where CX investment goes. Many organisations spend heavily on reducing average handling time, shortening queues, and compressing journeys. These are legitimate efficiency goals. But they do not, by themselves, improve remembered experience. A shorter journey with a weak ending and no emotional peak will be remembered less favourably than a slightly longer one that ends well.

This does not mean making journeys longer. It means recognising that the quality of the peak and the ending matters more than the duration of the middle. Effort spent on customer experience strategy that focuses exclusively on speed and friction reduction, without attending to the emotional architecture of key moments, is effort that will not show up in loyalty metrics.

The practical implication: when prioritising CX investment, map your journeys for emotional intensity first. Identify where the peak is currently occurring — is it a designed positive moment, or an accidental negative one? Then examine how every major journey ends. If the answer to either question is "we don't know," that is the starting point.

Applying the peak-end rule without manipulating customers

A reasonable concern about designing for memory is that it shades into manipulation — creating an impression of quality that the underlying experience does not merit. This is a legitimate ethical question, and it deserves a direct answer.

The peak-end rule is not a technique for deceiving customers about the quality of what they received. It is a recognition that the same quality of service, delivered with attention to emotional architecture, will be remembered more accurately than the same service delivered without that attention. A genuinely good experience that ends badly will be remembered as worse than it was. The goal of peak-end design is to ensure that memory reflects reality — not to inflate memory beyond it.

Where organisations cross the line is when they use peak-end thinking to paper over genuine service failures — engineering a warm ending to a journey that involved real harm, without addressing the harm. That is not CX design; it is reputation management in the wrong direction. The ethical application of the peak-end rule starts with fixing the actual problems, then designing the emotional architecture around a service that deserves to be remembered well.

What this means for how you measure and improve CX

The peak-end rule should change three things about how CX programmes operate.

First, measurement should identify emotional peaks, not just average scores. This means customer feedback management that captures the specific moments customers reference when they describe an experience — not just an overall rating. Text analytics, structured open-ended questions, and well-designed post-journey interviews all surface this. A number alone does not.

Second, journey design should treat the ending as a first-class design object. Every major customer journey should have an explicit ending design — not just a process close. Who delivers it, what it says, what emotional note it strikes, and how it connects to the next interaction should all be specified and tested.

Third, investment prioritisation should weight peak moments heavily. Not every touchpoint is equal in its contribution to memory. The moments that carry the highest emotional charge — whether by stakes, novelty, or vulnerability — deserve disproportionate design attention and resource. A framework for identifying and scoring these moments is more useful than an equal-weight optimisation across every step of the journey. Tools like Renascence's CX Maturity Assessment can help organisations understand where their current journey design is leaving these moments to chance.

The peak-end rule is not a trick. It is a description of how human memory actually works — and a standing challenge to any CX programme that optimises for the average rather than the memorable. The customers who stay, advocate, and return are not the ones who experienced the highest average. They are the ones who remember a moment that mattered, and an ending that felt right. Design for those two things, and the rest of the journey has a much better chance of mattering too.