Peak-End Rule: Which Moments Shape the Customer Experience?
The peak-end rule says that when we remember an experience, our minds do not add up every moment. Instead, they lean on the most intense moment and the final one. In this guide, we cover where the idea comes from, what it means for digital customer journeys, and which moments deserve your attention first. We are Talha Aslan and team, and we use this lens often in client work.
What is the peak-end rule and what does it explain?
The peak-end rule is a memory principle stating that people judge a past experience mostly by its most intense moment (the peak) and its final moment (the end), while largely ignoring how long it lasted. In other words, a memory is not an average of minutes. It is a summary of two moments.
This matters for customer experience. When customers decide whether to buy from you again, they do not replay every step. They use the summary they remember. So instead of trying to perfect every second, it pays to know which moments will shape that summary.
That said, the rule does not mean the middle is irrelevant. A bad middle can turn into a negative peak of its own. The rule simply predicts how the remembered evaluation forms.
Which studies does the peak-end rule come from?
The concept grew out of experiments by Daniel Kahneman and colleagues in the early 1990s. Barbara Fredrickson and Daniel Kahneman published "Duration neglect in retrospective evaluations of affective episodes" in the Journal of Personality and Social Psychology. Their participants rated short film clips afterward, and the length of each clip had a weak effect on those ratings.
Around the same time, Kahneman, Fredrickson, Schreiber, and Redelmeier published "When more pain is preferred to less: Adding a better end" in Psychological Science. The core idea is simple. Some participants preferred a slightly longer unpleasant experience when it ended more gently.
In addition, Redelmeier and Kahneman looked at a similar pattern in real life settings. We skip the details of that work here. For our purposes, the key point is that findings outside the lab also supported the weight of peaks and ends.
What is duration neglect?
Duration neglect means we barely account for how long an experience lasted when we look back on it. Fredrickson and Kahneman described it alongside the peak-end rule. The two ideas work like two sides of one coin.
For example, a customer may spend five or fifteen minutes on a shipment tracking page. Later, that difference often fades. On the other hand, the most frustrating moment or the most reassuring message in that time stays clear.
Therefore, "we made it faster" is not a strong improvement on its own. Speed has value, of course. However, if the remembered summary does not change, customers may not notice the gain. Our advice is to pair any speed work with deliberate design of the peak and end moments.
How do the lived experience and the remembered experience differ?
Kahneman often contrasts the "experiencing self" with the "remembering self." The experiencing self lives each moment. The remembering self writes the story. And the remembering self tends to make the decisions.
In digital marketing, this split has a concrete effect. When customers reorder or recommend a brand to a friend, they do not consult a minute by minute log. They consult the summary. As a result, repeat purchases, reviews, and referrals depend on the quality of that memory.
This view lines up with the decision stops in our film. In what is digital marketing, the "Make them feel" stop stands for the place where this memory forms in the customer's mind. When we design that stop, the peak-end rule is one of the most useful ideas we have.
How do you apply the peak-end rule to customer experience?
The first step is to break your journey into moments. Then you weigh each moment with two questions. Is this moment emotionally intense? Is this moment an ending? Moments that earn a yes on both are your priority.
- List your customer journey step by step, from first contact to after the sale.
- Next to each step, note the likely positive and negative feelings.
- Mark steps with high emotional intensity as peak candidates.
- Mark every point where an interaction ends as an end candidate.
- Plan small tests that reduce negative peaks and strengthen endings.
This list looks simple. Still, most teams never treat endings as a separate design task. For instance, the page that opens after someone clicks the buy button often stays a default template. Yet that page is the final moment of the shopping session.
Where do peak moments show up in a digital journey?
A peak is not always positive. Sometimes the most intense feeling is disappointment. So when you look for peaks, look at both delight and tension.
- The moment the product arrives and the box opens.
- An error message that appears during checkout.
- The moment a customer sees whether a discount code works.
- The first real reply from support.
- In software, the moment a user gets the first meaningful result.
For example, on a service website, the peak may be the moment a quote request goes out. The visitor feels expectation, curiosity, and a little worry. So the screen that follows carries both the peak and the end of the session. That is a rare chance, and we suggest you use it well.
Also, peaks vary by industry. In a furniture store, the peak may be assembly day. On a learning platform, it may be the first finished lesson. Measuring beats guessing here. Support tickets, session recordings, and reviews all point to intense moments.
Why do endings leave such a strong mark?
The ending decides how the story closes. The "adding a better end" study by Kahneman and colleagues showed that an ending can lift the overall rating. Therefore, leaving endings to chance means leaving the memory to chance.
A digital journey has many small endings, not one. A shopping session ends on the thank you page. An order ends with delivery. A support ticket ends with its closing message. Each one is the last frame of a separate memory.
On the other hand, do not think of an ending only as a happy close. The end of a return is also an ending. Even if customers send a product back, a fair and quick process can keep the relationship alive. In our field experience, brands that design the return experience tend to benefit. We share that as a general tendency, not as a percentage.
How does the thank you page shape the final memory?
The thank you page is the closing frame of a purchase or form session. On many sites, it only says "Your order has been received." Yet at that moment, the customer wonders one thing above all: what happens now?
So a good thank you page closes the uncertainty. It shows an order summary, an expected delivery window, the next step, and a reachable contact channel on one screen. That way, the session ends with clarity instead of worry.
This page also matters for measurement, since conversion tags usually fire here. When you redesign it, take care not to break your tracking. You can check your numbers with our conversion rate calculator. As for the physical card inside the box, we covered that in our ecommerce thank you card guide, so we will not repeat it here.
How can a confirmation email create a better ending?
The confirmation email is the lasting copy of the thank you page. Customers close the page, but they open the email again later. So this email works as a second ending that reinforces the memory.
We look at three principles in confirmation emails. First, clarity: what they bought, when it will arrive, and where to track it. Second, tone: a human voice instead of a corporate receipt. Third, easy access to help.
Example scenario: a home textiles store adds a short care tip for the product to its confirmation email. This small detail tells the customer, "We care about what you bought." Moreover, it adds no extra sales pressure. For the technical side of the template, see our HTML email design guide.
Why is delivery so often a peak?
For brands that sell physical products, delivery is the moment the wait ends. Customers finally get what they have looked forward to for days. That gives delivery both emotional intensity and a sense of closure.
However, delivery is often the step a brand controls least. The carrier, the delivery time, and the condition of the parcel sit outside your hands. Still, some things stay within your control: packaging, how the product sits in the box, what customers see first on opening, and how you communicate during delays.
Example scenario: a personal accessories brand notices a shipping delay and sends the customer a short update on its own. The customer hears about the problem from the brand before discovering it. As a result, a negative peak softens into "at least they kept me informed." This does not shorten delivery time, but it can change the story people remember.
How should you design support interactions around the peak-end rule?
Support usually starts with a problem. In other words, customers often arrive at support already carrying a negative peak. So two goals stand out: do not make that peak bigger, and close the ticket with a good ending.
- In the first reply, restate the problem in your own words so customers feel understood.
- Do not make customers repeat the same information again and again.
- After the fix, summarize what you did in a short closing message.
- In that closing message, state the next step and how to reach you again.
That said, the rule does not promise that a long support process with a good ending will feel flawless in memory. The rule describes a tendency. If wait time creates a serious negative peak, a good ending may not erase it. So we recommend improving both sides together.
How do peaks and endings look across channels?
The table below compares typical peak and end moments across digital touchpoints, along with the first step we recommend. It is a general framework. In every business, you should look at your own data first.
| Touchpoint | Likely peak | Likely end | Our suggested first step |
|---|---|---|---|
| Checkout and cart | Error message or discount code | Thank you page | Simplify error messages and add a clear next step to the thank you page |
| Confirmation email | Unexpected missing information | Closing line of the email | Move the tracking link and contact option to the top |
| Delivery | Opening the parcel | First use of the product | Design the unboxing and the delay notice |
| Support ticket | First description of the problem | Closing message | Write a closing template that includes a summary |
| Service quote form | Submitting the form | Screen after the form | Explain the expected response time and the process |
Notice that in some rows the peak and the end fall on the same moment. The service quote form is a good example. Such moments let one improvement capture two effects, so we suggest putting them at the top of your list. Use the table as a checklist and ask, for each row, "Did we design this moment, or is it a template?"
How do you find negative peaks?
Negative peaks are the friction points customers remember most strongly. To find them, follow the evidence instead of guessing.
The first source is support tickets. If the same topic keeps coming up, a peak likely sits there. The second source is session analytics. For example, if many users abandon one form step, that step may create tension. The third source is reviews. Emotional words in reviews show where in the journey feelings run high.
A usability review is a good way to run this process systematically. We explain how in our UX audit guide. For the broader picture of user experience, see our UX guide. That way, peak analysis becomes part of a wider experience review.
What should you watch for when you create a positive peak?
A positive peak can be a small moment that beats expectations. It does not need a big budget. A personal note, an unexpected usage tip, or fixing a problem before the customer notices can be enough.
However, we need to draw a line here. Creating a positive peak does not mean misleading people. A fake "only 2 left" warning, a countdown timer that is not real, or invented reviews are dark patterns. They may create a brief spike of excitement. Yet once customers see the truth, the memory turns negative.
These tactics can also carry risk under consumer protection rules. This is not legal advice, so please check with your own counsel before you act. Our principle is clear: you create a peak with real value, not with artificial pressure.
How does the peak-end rule affect loyalty and repeat purchases?
When customers decide to buy again, they consult their memory. According to the peak-end rule, that memory is a weighted summary of peaks and endings. So improving those moments can influence the path to a repeat purchase.
Still, we should be honest. No general rule expresses this link as a fixed rate. Every industry and every customer base behaves differently. That is why we suggest tracking the effect with your own data.
Two core metrics do the job. The first is churn rate, which we explain in our churn rate guide. The second is customer lifetime value, covered in our CLV guide. Compare both metrics before and after your improvements.
How does the rule combine with other behavior models?
This rule focuses on what happens after a decision: what do people remember once the experience ends? Other models look at what happens before a decision. Using both gives a fuller picture.
For example, the Fogg Behavior Model says a behavior happens when motivation, ability, and a prompt come together. We covered it in our Fogg Behavior Model article. A good ending can feed the motivation for the next purchase.
Similarly, the "messy middle" concept from Google's research team describes the loop between exploration and evaluation. You can read more in our messy middle article. When customers re-enter that loop, the memory of their last experience is an important input.
What are micro endings and why do they matter?
A micro ending is a small close inside a larger journey. Finishing one step of a form, closing a page, or reading a notification are all micro endings. Each one looks minor, but there are many of them.
For example, if account creation takes three steps, the end of each step is a small close. If users feel "I am on the right track" after each one, the whole experience feels smoother in memory. On the other hand, if each step ends on a vague screen, small worries add up and can become a negative peak.
So we check three small things at micro endings: clear feedback that the step is done, the number of steps left, and a way to go back. These elements tie closely to interface copy, which we discuss in our UX writing guide. In short, micro endings prepare the big ending.
Where is the ending in software and subscription products?
In subscription products, the experience flows without a clear break, so the ending is hard to spot. Still, there are natural closing points. These include the end of a session, the monthly invoice, the renewal reminder, and the cancellation screen.
Example scenario: a project management tool sends users a short summary of the work they finished each month. This summary shifts the monthly memory from "the invoice arrived" to "look what I got done." Moreover, it shows the product's value in concrete terms.
The cancellation screen needs special care. A cancellation flow made deliberately hard creates one of the strongest negative peaks. It may also fall under consumer rules in many countries; this is not legal advice. We suggest a short, respectful cancellation flow that leaves the door open. That way, even departing customers remember a decent ending.
When should you ask for customer feedback?
A feedback request is also a moment you design. Surveys and review requests touch the remembered summary of the experience. So timing matters.
For example, a review request sent before the product even ships feels early and pointless. On the other hand, a short question after customers have used the product a few times gathers more thoughtful answers. The same logic applies to support. A satisfaction question after the issue is truly solved gives you more meaningful data.
Also, the feedback form itself is an ending. A long, tiring survey can close a good experience on a note of fatigue. That is why we prefer short forms, one open question, and a closing screen that says thank you. Then use what you collect honestly, as we explain in our guide to using customer reviews in ads.
What are the limits and criticisms of the rule?
Like any finding in psychology, this rule has limits. Later research debated whether it holds with the same strength for every kind of experience. In some contexts, duration or the overall average can also shape judgments.
Also, the rule does not always define what counts as the peak or the end. In real life, it can be unclear where an experience stops. For a software subscription, is the end the close of a session, the monthly bill, or the cancel screen?
These debates do not cancel the concept, but they call for care. Instead of copying findings straight onto your audience, treat them as a hypothesis to test. The Nielsen Norman Group article on the peak-end rule in UX also frames it as a guide for design. In short, the rule tells you where to look, not what you will find.
Which mistakes should you avoid when applying the peak-end rule?
The mistakes we see most often come from misreading the rule. The list below sums them up.
- Polishing only the ending while ignoring big problems in the middle.
- Using dark patterns such as fake scarcity or fake timers to create a peak.
- Trying to turn every moment into a peak and wearing customers out.
- Breaking conversion tracking while changing the thank you page.
- Assuming a change "worked" without measuring it.
You can read this list as a quick check before each campaign or site change. Changes to thank you pages and email templates in particular can affect tracking and customer memory at the same time. For that reason, we plan each change as a small test. What these mistakes share is treating the rule as a trick. Instead, it is a map that shows where honest, careful treatment of customers makes the biggest difference.
Example scenario: how does the rule work on a service website?
Example scenario: an architecture studio collects quote requests through its website. A visitor fills in the form, clicks send, and the screen says only "Thanks." Then nothing happens for two days.
In this journey, the peak is the moment of sending the form. The ending is an uncertain wait. So the memory likely takes shape around the worry, "Did it even go through?"
The fix has three steps. First, the post form screen explains the process and the expected reply time. Second, an automatic confirmation email sends a summary of the request. Third, after the first call, the studio shares a short recap message. That way, every interaction closes with a deliberate ending. If you want to adapt this to your own forms, our web design service covers form and thank you flows together.
Is there a practical checklist for the peak-end rule?
Yes. You can bring the concept into weekly work with a short checklist. It draws on the steps we have seen make the most difference in practice.
- Does your thank you page clearly state the next step and a contact option?
- Does your confirmation email show tracking details and help options near the top?
- When delays happen, do you inform customers first, or do they have to ask?
- Do your support tickets end with a closing message that includes a summary?
- Have you identified the recurring negative peak in the last three months of support topics?
- Do you track the effect of your changes with a metric?
If you want to extend these steps across your whole order flow, our ecommerce consulting work reviews delivery, returns, and communication together. Put simply, the goal is to stop leaving remembered moments to chance.
Core guide: What Is Digital Marketing? Its types, examples and how it is done, on one page with an interactive film that travels through the brain.
