• long Read

Stumbling on Customer Service

Some of the most useful books for business aren’t about business at all. “Stumbling on Happiness” by Daniel Gilbert is a discussion of how people actually anticipate, perceive, and recall their experiences.

Some of the most useful books for business aren’t about business at all.

Recently, I read a terrific book that is extremely readable and very relevant to business, Stumbling on Happiness, by Daniel Gilbert. Gilbert is a professor of psychology at Harvard. This book is a very well-written and well-researched discussion of how people actually anticipate, perceive, and recall their experiences.

Bottom line: many of the assumptions that we make about how customers will react to customer service and many other customer-facing activities are misleading and incorrect. Consequently, we have a big opportunity to shape our customers’ experiences in ways we probably haven’t imagined. Similarly, we make systematic misjudgments in our decision-making, both in daily activities and in major initiatives.

I strongly recommend reading this book. I offer below a number of interesting points–all well documented–that emerge from the book. I’ll make a few suggestions about how these points tie into business activities, but leave it up to your imagination to draw the inferences for your business.


We tend to overestimate the likelihood of good events occurring; thus people are unrealistically optimistic. Even if a bad event occurs (e.g. an earthquake), within a few weeks people are unrealistically optimistic again. (Think about the implications for deciding whether to end an initiative or program with only mediocre performance.)

People are even more unrealistically optimistic if a very bad event occurs. For example, cancer patients are more optimistic than their healthy counterparts.


If a bad event is predictable and we anticipate it, we develop ways to rationalize and adjust to the impact. This gives us a feeling of control. Gilbert cites experiments in which subjects who receive strong shocks in a regular pattern deal with it better than subjects who receive mild shocks at unpredictable intervals. (Think about your policy for letting customers know in advance about service problems.)

People find it gratifying to exercise control, not for the result but for the exercise itself. When people lose the ability to control things, they feel unhappy, helpless, and despondent. Losing control has much worse effects (on health and wellbeing) than never having had control. The feeling of control is one of the wellsprings of mental health. (I recall reading about classic industrial engineering experiments in which assembly line workers who were allowed to work to the bottom of a bucket of bolts were much happier and more productive than those who had the bucket endlessly refilled.)

We really value the freedom to choose. People will pay a premium today for the opportunity to change their minds later, even if the economic consequence is overwhelmingly unfavorable.

Shaping perceptions

If a person describes his or her reaction to an event just after the event, the person will recall the description rather than the event, even if the description is not accurate. Most events have both positive and negative aspects. This means that you can shape someone’s recollection by asking questions that highlight the aspects you want the person to remember.

This works prospectively as well. If we suggest that someone focus on one aspect of an event, he or she will tend to disregard the other aspects. There are several commonly used film clips that illustrate this. For example, in one clip, viewers are asked to count the number of times a group of people pass a ball to one another, and the viewers fail to notice a gorilla walking through the group. (Again, this suggests ways to shape customers’ perceptions of actual experience.)

Information acquired after an event actually alters a person’s memory of the event. (For example, if you give a report card on your overall service, it will obscure a customer’s recollection of individual problems.)


When we imagine an upcoming event or an object (e.g. a plate of spaghetti), we visualize it as being much richer and more ideal than the full range of possibilities. This inevitably leads to disappointment.

People have a general inability to think about the absence of events. For example, a person will readily remember tripping, but not be aware of all the times he or she didn’t trip. (This is why rare customer service problems unrealistically dominate customers’ perceptions.)

When we are selecting something, we focus on the most positive attributes. When we are rejecting something, we focus on the most negative. This leads to a common situation in which a person will both select and reject the same thing at the same time by focusing alternatively on positive and negative attributes, becoming paralyzed with indecision. Here’s another example of our tendency to overestimate the extremes: we overestimate the happiness of people in California, and underestimate the happiness of people with chronic illness or disabilities–both have a “normal” degree of happiness.

Predicting the future

When we think about events in the distant future, we focus on why they will happen; when we think about events in the near future, we focus on how they will happen. Think about promising to babysit for a weekend several months in the future versus thinking about the details of what you will do when the weekend starts tomorrow. (This is very relevant to evaluating mergers and other big initiatives: from a distance we leave out consideration of the implementation details, and later wonder why we committed to do it. The upshot: get immersed in the details early so your evaluation is realistic.)

People tend to assume that the future will look largely like today. Underestimating the future is a time-honored tradition. Arthur Clark observed, “When a distinguished but elderly scientist states that something is possible, he is almost certainly right. When he states that something is impossible, he is very probably wrong.” People almost always err by predicting that the future will be too much like the past.

When we have “holes” in our conceptualization of the past and the future, we plug in today. This means that our predictions are prejudiced by assumptions that things will not change. It also means that our view of past relationships and events is changed by what we feel today. This is especially powerful in predicting how we will feel, or in remembering emotions. People have great difficulty imagining that they will later (or previously) feel differently than today. This is not a “logical” issue, but a psychological one: we can visualize objects changing, but can’t “prefeel” emotions differently from our current experience.

We remember feeling what we believe we must have felt, rather than what we actually felt. Memory is less like a photograph than a collection of impressionistic paintings.

Preference for variety

People imagine that they would prefer variety more than they do. The reality is that they only prefer variety if habituation sets in. Otherwise, in episodes separated by time, they strongly prefer favorites.

Relative preferences

People are much more sensitive to relative amounts than to absolute amounts. In experiments, subjects preferred to receive a pay progression of $30 followed by $40 followed by $50, rather than $60 followed by $50 followed by $40, even though the latter amounted to more. Similarly, we prefer the sequence of a bad deal changed to a decent deal, to the sequence of a great deal changed to a good deal–even though the latter is “objectively” better.

We are more willing to contribute a small amount if we are first asked to contribute a much larger amount. The same applies to paying for a product or service. This is why stores often place an overpriced item next to a very overpriced item.

It is very hard for people to choose among similar items. In experiments, physicians faced with the choice of prescribing two similar medications prescribed neither much more often than those physicians who only had one medication available. (Think about the implications for product assortments.)

People prefer to avoid a loss, more than to have the possibility of a gain. Many transactions fail because the seller overestimates his or her prospective unhappiness, while the buyer underestimates his or her prospective happiness. The seller implicitly wants compensation for a possible powerful loss, while the buyer expects a less powerful gain. (Think about the implications for negotiating an acquisition.)

Perceptions of difficulties

People are very resilient in the face of trauma. Most bereaved people experience a relatively short period and low level of distress, relative to their expectations. The same is true of people who suffer a disability.

In fact, most people consider their lives enhanced by the experience of trauma. This is the result of what Gilbert calls our “psychological immune system,” which causes us to adjust quickly and rationalize the impact of a truly traumatic event. In a strange example, able-bodied people are willing to pay much more to avoid being disabled than disabled people are willing to pay to be able-bodied again. When things really go wrong, our minds actively shape our perceptions to make the best of things.

Paradoxically, it is more difficult to achieve a positive view of a somewhat bad experience than a really bad experience–because in truly difficult circumstances our minds find ways to justify and rationalize the new situation, while they do not do this with normal setbacks. Suffering triggers defense mechanisms that eradicate it, but we can’t predict that this will happen.

We are more likely to develop a positive view of something if we are stuck with it. Hence, patients feel more distress if a medical test is inconclusive than if it is positive.


We seek data that confirms what we already believe. In studies, researchers have shown that the wide availability of information on the Internet has led to increased polarization, as people focus on the information that reinforces their preconceptions, rather than weighing information that supports alternative positions.

Similarly, our minds select from our memories the “facts” that fit our preconceptions, and this strongly shapes our perceptions and even our recollections.

People ask questions that elicit the answers they want to hear.

Interestingly, when people have a problem, they especially seek information about others doing more poorly than themselves. For example, in a study, 96% of cancer patients said that they were in better shape than the average cancer patient.

Expectations and regret

When we feel dread in anticipation of something, we assume that this is how we will experience and recall the event. Our most consequential choices (e.g. marriage, profession) are most shaped by how we imagine our future regrets. We are especially prone to exaggerate our view of future regret when the choice is unusual rather than more conventional. (“Showcase” projects can ameliorate this because they are a low-risk way to generate new experiences.)

In an interesting study, 90% of respondents expected to feel more regret if they foolishly switched stocks, than if they foolishly failed to switch stocks. Most people think they will regret foolish actions more than foolish inactions. But in reality, after the fact 90% of people really regret not having done things more than they regret things they have done. It is easier to visualize or imagine inactions than new actions with uncertain consequences.

People consistently choose certainty over uncertainty, and clarity over mystery.

The bottom line: it is very hard to predict accurately our reactions to future events because we can’t imagine them, or what we will think and feel if they happen.

Changing the perception of experiences

Explanations change people’s extreme perception of experiences. For example, simply talking, and especially writing about a trauma will generate surprising improvements in well being and physical health. This is particularly true when the writing contains an explanation of the trauma. Similarly, writing ameliorates the impact of an extremely good event.

What is memorable

We remember unusual and infrequent events (e.g. where we were on 9/11) and assume that they are more common than they really are. (Perceptions of customer service are formed by the worst experiences, not the average.) This is why we repeat mistakes so often.

We remember the end of a sequence much more than the beginning, middle, or average. People’s perceptions are overwhelmingly shaped by the most recent events. In studies, people were more concerned by how they would feel at the end of their lives, than about the total amount of happiness they experienced in their lives.

Making predictions

The best way to make predictions about future happiness is to find someone who is now having the experience and ask how he or she feels. (This is especially important in a job search. It also reinforces the importance of “showcase” projects.)


There are three main problems with the way our imagination works: (1) our minds fill in and leave out information in predictable, but unexpected ways, and we don’t even realize it; (2) we project the present into the future, and again don’t realize it; and (3) we fail to see that things will look different once they happen.

Here’s a final thought: the average person doesn’t see himself or herself as average. We attribute other people’s choices to features of the chooser (e.g. Jim prefers red ones), while we attribute our choices to characteristics of the choice (e.g. it had a richer color).

Customer service and decision-making

When we conceptualize and measure customer service, we most often simply assume that people are “rational.” Thus, it seems obvious that if we produce 96% service, it is better than 95% service.

But the insights that Gilbert offers instruct us that there is a lot more to it than that. A simple example: If the 95% is accompanied by a warning and explanation of impending problems, and the customer can choose a substitute or alternative, the customer will be happier than if he or she received a higher nominal service level with no “heads-up” or ability to exercise control. (Remember: the only thing worse than bad news is no news.)

Similarly, a company with a 98% service level with a few memorable disasters will be perceived more negatively than a competitor with a 94% service level accompanied by report cards reminding the customer of the “great” overall service and a few anecdotes of “heroic” service incidents.

Stumbling on Happiness also sheds light on a number of systematic errors and misperceptions that we naturally make in evaluating decisions and predicting how we will react to future events.

These factors are critical to the way we conduct our daily business, as well to the processes we use to frame and evaluate our major strategic moves. If we are aware of these tacit factors, we can take deliberate steps to significantly improve our performance. The key is awareness and understanding, and all at no cost.

Reprinted from IslandsofProfitBook.com

Jonathan Byrnes is the author of Islands of Profit in a Sea of Red Ink. Byrnes, a Senior Lecturer at MIT, is an acknowledged authority on profitability management. His experience spans virtually every industry, including health care, transportation, software, retail, financial services, distribution, and others. He is President of Jonathan Byrnes & Co., a consulting company founded in 1976. In that role he has advised over fifty major companies, medical institutions, and industry associations.

About the author

Jonathan L.S. Byrnes, Senior Lecturer at MIT, is an acknowledged authority on profitability management.



More Stories