Your guest doesn't evaluate their visit as the sum of all its moments. They judge it by two points: the most emotionally intense moment and how it ended. You have full control over one of them. You're probably not managing it.
The discovery
In 1993, Daniel Kahneman — the psychologist who would receive the Nobel Prize in Economics in 2002 for his work on human decision-making — published with his team the results of an experiment with serious implications for every industry in which human experience matters. Hospitality and hotel operations belong at the top of that list. And yet this body of knowledge has reached the HoReCa sector in remarkably limited form.
Kahneman and colleagues discovered that people do not evaluate a completed experience by calculating its "average" — summing all the moments and dividing by the duration. Instead, the brain applies a radical shortcut: it evaluates the experience almost entirely through two points. The first is the emotional peak — the most intense moment in the entire experience, positive or negative. The second is the ending — the final moments before leaving the situation.
Everything in between — including total duration — has surprisingly little influence on the final rating and on what stays in memory. Kahneman called this the peak-end rule.
The experiment
The original Kahneman experiment concerned pain. Participants were asked to submerge a hand in ice-cold water under two conditions. In the first: 60 seconds in water at 14°C. In the second: those same 60 seconds, after which the water temperature rose slightly for a further 30 seconds — still cold, but marginally less intense.
Objectively, the second condition was longer and contained more total discomfort. Subjectively, participants consistently rated it as less painful and more frequently chose it for repetition. The ending was the deciding factor — the less intense final moments rewrote the memory of the entire experience.
Redelmeier and Kahneman (1996) replicated this effect in a medical setting — during painful colonoscopy procedures. Patients whose procedures were extended by a less intense final phase recalled them as less painful in retrospect — even though total duration and discomfort were greater. The ending rewrites the memory.
Here is what that means for hospitality.
The Peak-End rule
Think of a typical restaurant visit as a sequence of moments: the greeting at the door, being seated, the first look at the menu, the first contact with the server, ordering, waiting, the starter, the main course, conversation at the table, dessert or coffee, asking for the bill, waiting for the bill, payment, farewell, departure.
Every moment contributes something to the experience. But according to the peak-end rule, what the guest will actually remember — and what will determine whether they return, recommend the venue, or write a five-star review — is dominated by two things: the highest-intensity moment of the evening and the last few minutes.
What do the last few minutes of most restaurant visits look like? The guest has finished eating. They sit at the table, perhaps with an unfinished coffee. The server clears the plates. The guest asks for the bill — or waits until someone notices. A card terminal appears. Tap or pin. Someone says "thank you, please come again" without making eye contact, because they just took an order from another table. The guest stands, collects their coat, and leaves.
That sequence — largely rushed, transactional, and impersonal — is the last thing the guest's brain encodes before forming its final assessment of the entire evening. The extraordinary dish. The perfect wine pairing by the sommelier. The attentive and warm service during the main course. All of it is filtered through that final sequence. And in many cases, that sequence quietly reduces the rating the guest hasn't yet consciously formulated.
Where the peak is?
Most hospitality operators assume that the emotional peak of a guest's visit is the main course. This assumption is frequently wrong.
The peak is the moment of highest emotional intensity — which might be the genuine surprise of an amuse-bouche, the moment a server remembered a preference from a previous visit, a gesture at a birthday table, or — and this is the most consistently underappreciated category — a moment of authentic human connection that had nothing to do with the food at all.
Fredrickson and Kahneman (1993) showed that the peak moment doesn't need to be long. It can last a few seconds. But it needs to be real — not scripted, not read from a service checklist. Guests are remarkably sensitive to inauthenticity, and an engineered "wow moment" that is detectable produces the opposite effect from the one intended.
This tells experienced hospitality operators that the peak cannot always be planned in advance. But the conditions for it can be created: a team that is present, properly informed, and has the latitude to respond to what is actually happening at the table, rather than executing a predetermined sequence.
Your guest doesn't give you a star rating for the average of their evening. They give it for the peak and the end. One of those you can design. Both of them you can ruin.
Solution
The ending is entirely within the operator's control. And it is almost universally underinvested.
The Redelmeier and Kahneman (1996) evidence shows that even in the context of pain — an objectively negative experience — adding a gentler final phase significantly improved the retrospective rating of the whole. The ending rewrites memory in the positive direction. In hospitality, where the experience is inherently positive in nature, this effect is stronger and more readily accessible.
What can be done concretely? Below are several principles that require no budget — only intention.
Manage the bill-waiting moment. The moment when a guest asks for the bill and waits is one of the most common sources of end-of-visit frustration. Not because the wait is long — but because it happens without any signal that someone has noticed. A simple acknowledgement ("I'll bring it over in a moment") produces a disproportionate effect relative to the effort.
Personalise the farewell. A server who, in saying goodbye, references something specific from the evening ("I hope the dessert was worth it") leaves a different trace than one who says "thank you" to the space above the guest's shoulder.
Create a physical final moment. Depending on the venue's positioning, this might be something small with the coffee at the end, a card, a printed recipe, anything the guest takes with them as a physical carrier of a good ending.
Consider the external farewell. If the venue has a cloakroom or a staffed entrance, the moment at the door carries weight disproportionate to its duration. The exit interaction — whoever delivers it — holds an outsized share of the evening's final rating.
Implications
If the peak-end rule governs how guests remember and rate their experience, then the most operationally significant question in hospitality is not "how do we maintain consistent food quality?" — though that matters enormously. It is: "what is the emotional peak of our guest experience, and what does our ending actually feel like?"
Most hospitality operations have detailed protocols for the arrival sequence and table service. Almost none have a deliberate protocol for the end of the visit. This is not a minor oversight. It is a persistent leak in the guest experience that no amount of culinary excellence can fully compensate for.
The good news: the ending is inexpensive to fix. It requires no new equipment, no additional budget, no menu redesign. It requires attention, intention, and a team that understands why those last three minutes matter more than most managers have ever told them.
References
Kahneman, D., Fredrickson, B. L., Schreiber, C. A., & Redelmeier, D. A. (1993). When more pain is preferred to less: Adding a better end. Psychological Science, 4(6), 401–405.
Fredrickson, B. L., & Kahneman, D. (1993). Duration neglect in retrospective evaluations of affective episodes. Journal of Personality and Social Psychology, 65(1), 45–55.
Redelmeier, D. A., & Kahneman, D. (1996). Patients' memories of painful medical treatments. Pain, 66(1), 3–8.
Kahneman, D. (2011). Thinking, fast and slow. Farrar, Straus and Giroux.
Pine, B. J., & Gilmore, J. H. (1999). The experience economy. Harvard Business School Press.
Monday - Friday 9 -17
Shrewsbury
United Kingdom
07925603011
baldhospitality@gmail.com