How to Improve Customer Satisfaction: Why Measuring Isn't Enough
Most organisations measure customer satisfaction obsessively. NPS surveys after every interaction, CSAT scores by department, mystery shoppers, quarterly research programmes. The dashboards are full of data. And yet scores barely improve structurally. The same pain points surface in reports year after year.
Something is fundamentally wrong with the approach. Not with measurement itself, but with what happens afterwards. Or more precisely, with what doesn't happen.
Why doesn't more measurement lead to better customer satisfaction?
NPS and CSAT are useful instruments, but they have a fundamental problem: they measure stated preference rather than revealed preference. Customers report what they think they feel at the moment of the survey. But what they actually remember, and what drives their future behaviour, is something different.
Consider the gap between what someone says about a restaurant visit and whether they actually return. The two correlate far less than we expect. People are poor predictors of their own future behaviour, and the questions we ask in customer satisfaction research are part of that problem.
The second issue is that NPS is treated as a driver rather than an outcome. Organisations hold their teams accountable for NPS improvement as if the score itself can be changed. But NPS is the result of experiences, not the cause. You can no more directly improve an NPS score than you can lower your blood pressure by adjusting the gauge.
The third problem is regression to the mean. Journeys are optimised touchpoint by touchpoint, and the effect of those optimisations gets averaged out in headline scores. But customers don't experience averages. They remember peaks and endings.[1]
How does customer memory actually work?
Daniel Kahneman's research into pain perception uncovered something that should radicalise the entire CX field: people do not remember an experience as the average of all its moments. They remember it based on two things: the emotional peak and the ending.[2]
This principle, the peak-end rule, has been confirmed in dozens of studies. In one of the most cited experiments, Redelmeier and Kahneman had patients undergo two versions of a painful medical procedure. The shorter version was objectively more painful. Yet patients rated the longer version as less unpleasant, simply because it ended on a less intense moment.[3]
The implications for customer experience are profound. The duration of an interaction barely matters. Most individual touchpoints barely matter. What matters is this: how did it feel at the most emotionally loaded moment? And how did it feel at the end?
A customer who waits thirty minutes with customer service but is helped exceptionally well will remember the interaction positively. A customer who gets an unsatisfying answer in ten minutes will remember it negatively, regardless of how friendly the agent was for the rest of the call.
System 1 makes the judgement
The second driver behind this mechanism is Kahneman's dual-process theory. We have two thinking systems: the fast, automatic System 1, and the slow, deliberate System 2. Customer satisfaction is largely a System 1 judgement. It is not formed through rational evaluation of all individual touchpoints, but through an emotional impression that forms automatically based on a few critical moments.
This also explains why information campaigns about customer experience deliver so little. You can train staff in friendliness, improve procedures and manage expectations, but if the emotional architecture of the customer journey is wrong, none of that translates into higher satisfaction scores.
The SUE Influence Framework as a customer experience analysis tool
So how do you analyse what is really happening in the experience of your customers? The answer does not lie in better surveys, but in a deeper understanding of the psychological forces that drive customer behaviour and perception.
The SUE Influence Framework maps those forces. For each touchpoint in the customer journey, it asks five questions:[4]
- Job-to-be-Done: What is the customer actually trying to achieve at this point in the journey? Not the functional task, but the deeper motivation behind it.
- Pains: What frustrates the customer in the current experience? What makes it hard, uncertain or disappointing?
- Gains: What would genuinely delight the customer? Which benefit are they seeking that is currently absent?
- Anxieties: What fears and doubts are present? What makes the customer uncertain about the next step?
- Comforts: What habits or expectations keep the customer locked into current behaviour? What makes change difficult?
This is fundamentally different from a traditional journey map. A journey map tells you what customers do at each touchpoint. The SUE Influence Framework tells you why they do it, and which emotions and cognitions are at play.
Designing a customer experience that genuinely scores 9+ starts with understanding the psychology behind it.
In the Deep Dive Designing the 9+ Customer Experience, you learn how to apply the SUE Influence Framework, the peak-end rule and practical design principles to structurally improve your customer experience. 12 online lessons, €690 excl. VAT, lifetime access.
How do you design customer satisfaction in practice?
Customer satisfaction is an outcome, not a driver. You cannot improve it directly. You can only design the experiences it comes from. Four principles guide that work.
Principle 1: design the emotional peak deliberately
Identify which moment in the customer journey carries the greatest emotional weight. This is often not the moment you would expect. It is not always the purchase, the onboarding or the complaint, but the moment at which the customer feels most vulnerable, uncertain or excited.
Invest disproportionately in that moment. A positive emotional peak at the right point has more impact on remembered satisfaction than dozens of improvements to functional touchpoints that no one really registers.
Principle 2: end on a high
The ending of an interaction is remembered disproportionately strongly. A complaint that is resolved well, an order that arrives with a handwritten note, an offboarding that feels smooth: these closing moments colour the overall judgement of the experience.
Actively design what the customer experiences last in every type of interaction. Many organisations invest heavily in acquiring and onboarding customers, yet leave the endings of individual contact moments to chance.
Principle 3: remove anxieties before the next step
Anxieties are the most underestimated force in customer experience. Customers who are uncertain about what will happen after a purchase, a complaint or a contract renewal rate their experience lower, even when the actual experience was good.
Proactive communication about what comes next, transparency about processes and social proof at critical moments reduce those anxieties structurally. Not by providing more information, but by providing the right information at the moment when doubt is greatest.
Principle 4: reduce friction at decisive moments
Friction costs emotional energy. And emotional energy that a customer must invest in getting something done comes directly at the expense of their perceived satisfaction. Long waiting times, unclear forms, multiple transfers between agents: customers feel this as effort, even when it is functionally acceptable.
Kahneman's research shows that the effort people must exert to achieve something significantly reduces their appreciation of the result. Simply halving the friction at your three most used contact points has more effect than a complete new service training programme.
Frequently asked questions about improving customer satisfaction
Why doesn't customer satisfaction improve despite extensive measurement?
Because most measurement tools record what customers say, not what they feel and remember. Customer satisfaction is an emotional memory formed by a few decisive moments, not by the average of all touchpoints. Measuring without understanding that underlying psychology results in treating symptoms rather than causes.
What is the peak-end rule and why does it matter for customer experience?
The peak-end rule is a cognitive principle discovered by Kahneman: people remember an experience not as the average of all its moments, but based on the emotional peak and the ending. Organisations that invest disproportionately in the most emotionally loaded moment and close each interaction on a strong note will realise structurally higher customer satisfaction.
Is NPS useless then?
No, NPS is useful as a diagnostic instrument. The problem is not the measurement, but the management approach to it. NPS is an outcome, not a driver. Managing to NPS without knowing which experiences cause it is symptom treatment. NPS becomes valuable when you combine it with qualitative research into the emotional architecture of your customer journey.
How do you use the SUE Influence Framework for customer satisfaction?
The SUE Influence Framework analyses the five psychological forces at each touchpoint: Job-to-be-Done, Pains, Gains, Anxieties and Comforts. By looking at each pain point in the customer journey through this lens, you discover where real dissatisfaction lives and where the greatest improvement potential lies. Read more in The SUE Influence Framework Explained.
What is the difference between measuring customer experience and improving it?
Measuring records an outcome. Improving requires understanding the underlying causes. Customer satisfaction is an emotional memory, not a sum of functional ratings. To improve it structurally, you must design the experiences it comes from: the emotional peaks, the closing moments, the removal of anxieties and friction at decisive points.
Conclusion
Improving customer satisfaction is not a measurement challenge. It is a design challenge. The science is clear: customers do not remember averages, they remember emotional peaks and closing moments. They do not act on the basis of rational evaluations of every touchpoint, but on the basis of the emotional memory formed by an experience.
That means investment in better dashboards and more measurement reaches diminishing returns at some point. The real gain lies in understanding the psychological forces that shape customer experience, and then redesigning the experiences that carry the greatest emotional weight.
Want to work on this concretely? The Deep Dive Designing the 9+ Customer Experience gives you the tools, the frameworks and the practical knowledge to do exactly that.
1.5 minutes of Influence
Read by 10,000+ professionals · Free · Unsubscribe any time