Share article
Try to recall the last flight you took. You probably don’t remember the check-in queue minute by minute, or the exact wording of the announcements, or how many times the drinks trolley came down the aisle. What you probably do remember is one moment when things were either brilliant or terrible, and how you felt walking off the plane. The rest is gone, dissolved somewhere in time.
This isn’t a coincidence, and it isn’t a memory glitch either. It’s simply how the human brain processes experience, and it’s one of the most well-documented phenomena in behavioural psychology. In customer experience (CX), it has gradually become one of the most practical tools we have, provided we know how to use it.
We forget processes. We remember feelings
When companies map the customer journey, they tend to draw it as a sequence of steps: sign-up, order, delivery, complaint, renewal. It’s tidy, it presents well to leadership, and KPIs slot neatly into it. The trouble is, that’s not how customers actually experience it. They don’t go through a process, they go through a series of emotional states, and it’s those states that become the brand’s lasting imprint in their memory.
Long-running Forrester research backs this up. Forrester measures CX quality across three dimensions: effectiveness (did the customer achieve their goal?), ease (how much effort did it take?), and emotion (how did they feel?). When they first compared which of these three dimensions had the biggest impact on loyalty back in 2014, emotion won. A repeat analysis on 2015 data confirmed it: emotion was the strongest predictor of loyalty in seventeen out of eighteen industries studied. In its 2024 report, Forrester showed that elite brands (the top 5% across industries) generate an average of twenty-five positive emotions for every negative one among customers, while the average company fares considerably worse on this front. And companies Forrester labels “customer-obsessed”, meaning those that genuinely put the customer at the centre of decision-making, grow revenue 41% faster, profit 49% faster, and retain customers 51% better than those that don’t.
That’s a figure worth holding onto. Not because emotion is a nice extra to have, but because without it, your CX strategy is measuring something other than what actually drives decisions.
How to map the emotional curve without just making it up
Here’s something I had to learn the hard way myself: emotional mapping easily degenerates into five marketing people sitting round a whiteboard, sketching a curve based on what they assume the customer feels. It’s tempting, it’s quick, and unfortunately it’s almost always wrong. Emotions aren’t something you can estimate from your desk. You have to measure them where they actually happen.
Effective emotional mapping usually combines three layers of data. The first is in-the-moment feedback, a short prompt right after a specific interaction (typically CES, Customer Effort Score, a metric measuring how much effort the customer had to put in to get something resolved), paired with a simple emotional scale: how do you feel right now, from frustrated to delighted. The second layer is qualitative: interviews and open-ended responses where the customer describes, in their own words, what they experienced at that point and why. A number tells you the emotion dropped. An interview tells you whether it dropped because of waiting time, the agent’s tone, or because the customer felt stupid having to ask something that should have been obvious. Three completely different causes, three completely different fixes. The third layer is behavioural: what the customer actually does, where in the process they abandon their basket, where they call the helpline instead of finishing a step online, where they backtrack.
Only once you overlay these three layers onto a timeline of the journey’s individual stages does something worthy of the name “emotional curve” emerge. And suddenly you see things a process map never revealed, like the fact that a stage you’d internally dismissed as trivial (a confirmation email, waiting for an SMS code, the wording of an error message) is actually the most emotionally sensitive point in the entire journey.
Why peaks and endings decide everything, not averages
This is where we get to the heart of the matter, historically tied to the name Daniel Kahneman, the Israeli-American psychologist and Nobel laureate in economics. In the 1990s, Kahneman, together with colleagues Barbara Fredrickson, Charles Schreiber and Donald Redelmeier, described what’s known as the peak-end rule: how someone recalls and evaluates a past experience is determined almost entirely by two points, the most intense moment of the experience and the moment at its end. Everything else, the whole long middle stretch, largely fades from memory.
The best-known demonstration is Redelmeier and Kahneman’s 1996 study, which tracked real pain ratings from patients undergoing colonoscopies. They found that patients’ retrospective assessment of how painful the whole procedure was barely correlated with its length at all, but correlated strongly with pain intensity at its peak and in the final three minutes. Some patients had procedures that lasted tens of minutes longer than others, yet didn’t rate them as worse in hindsight, as long as the ending was calmer. In an earlier 1993 experiment, participants held a hand in extremely cold water, and when given a choice between a shorter version of the ordeal or a longer one that ended in a slightly less painful phase, most (around 80%) chose the objectively longer suffering, because it was “stored” in memory as milder.
What does this mean for CX? That the overall length and smoothness of the customer journey matter far less than we assumed. You can have a process that’s 90% flawless and still have the customer rate it poorly if the peak was frustrating or the ending unpleasant. Conversely, a process with one rough moment in the middle but a strong, gracious ending can be rated surprisingly well in hindsight. This isn’t manipulation, it’s simply respecting how human memory actually works.
In practice, this means that when deciding where to put a limited CX improvement budget, the peak and the ending of the journey deserve priority far beyond their share of the total number of steps. The final email after a complaint is resolved. The last line in a support chat. The moment an order is collected. These are the points that decide the memory trace you leave behind.
Frustration as an opportunity, not just a loss
This point still catches me off guard whenever I revisit the CX literature, even after years of knowing it. There’s a phenomenon called the service recovery paradox, first formally described by Michael McCollough and Sundar Bharadwaj in 1992, building on the earlier Harvard Business Review article “The Profitable Art of Service Recovery”. The idea is simple, and a little counterintuitive: a customer who experiences a service failure that the company then fixes exceptionally well can end up more loyal than if the failure had never happened at all.
The psychological reason is straightforward: a successful recovery gives the customer concrete proof of how the company behaves when things go wrong, and that proof builds trust far more powerfully than a smooth ride where no test ever occurred. But simply solving the problem isn’t enough on its own. It has to genuinely satisfy the customer, it has to come with sincerity, and above all, it can’t happen again. More recent research on the topic points to an important limit: the recovery paradox only holds reliably for a one-off failure that’s handled well. If the same mistake repeats, or if the fix itself fails (a so-called double deviation), the effect doesn’t just disappear, trust collapses far faster than if the first mistake had never happened.
In practice, this means rethinking how companies view complaints. We often treat them as a necessary evil, a cost to minimise and close out as fast as possible. But this is precisely the moment when the customer is emotionally most vulnerable, and it’s also the moment with the greatest potential to shift the whole emotional curve upward, provided it gets the attention it would otherwise never receive. The point isn’t to encourage companies to go looking for mistakes. It’s to stop treating the moment of recovery as an administrative chore and start seeing it as an investment opportunity with one of the highest possible returns anywhere in the customer journey.
Connecting it to the hard numbers
The most common objection I hear from CX teams is: emotion is a lovely theory, but how do I defend it to the CFO? The answer is that the emotional curve isn’t in opposition to hard metrics, it’s what explains them.
NPS (Net Promoter Score, a metric measuring how willing a customer is to recommend a company) tells you how many people would recommend you. It doesn’t tell you why. But place NPS alongside an emotional map of the same journey, and you’ll typically see that the score drop lines up exactly with the point where the emotional curve dips lowest, and that the reason people leave despite a technically flawless process is almost always emotional: feeling ignored, feeling a loss of control, feeling treated like a ticket number rather than a person. Forrester’s data confirms this at a macro level: effectiveness and ease have stagnated or even declined at many companies in recent years, but emotional quality remains what separates the companies that grow from those merely surviving.
For a CX team, this points to one practical recommendation: always pair transactional metrics (CES after an interaction, CSAT after a resolved ticket) with a simple emotional question, and pair relational metrics (NPS, retention rate, customer lifetime value) with an emotional audit of the journey’s key peaks and endings. Only this combination lets you not just see that something’s wrong, but understand why it’s wrong, and crucially, where to invest the team’s limited energy so it actually shows up in the numbers leadership reads.
Customers will never thank you for a flawless process they don’t remember. They’ll stay loyal to you for the one moment they felt genuinely good, exactly when it mattered most.
