When Customers Complain, Listen to the Emotion

When Customers Complain, Listen to the Emotion

Dr. George Monray on what 1,633 Capital One customer reviews revealed about frustration, distrust and the psychology of financial complaints

By Inderscience.

A customer complaint is usually treated as a problem to be resolved.

For Dr. George Monray, it can also be something else: data.

In 2015, the business researcher presented research at the Society for Interdisciplinary Business Research conference in Bangkok examining the feelings, sentiments and perceptions expressed by dissatisfied customers of Capital One credit cards in the United States.

Rather than relying on a conventional customer-satisfaction survey, Monray analyzed 1,633 written consumer reviews published on ConsumerAffairs between May 2009 and January 2015. Using qualitative data analysis and inductive coding, the research identified 38 categories of negative issues and six principal emotional responses.

The results offered a different way of looking at customer complaints.

Frustration, extreme frustration and anger represented 38% of the coded feelings. Perceptions that the company was acting fraudulently or had breached its contract represented another 24%, while the perception that the institution was hiding from customers or failing to confront problems accounted for 15%.

Together, those three categories represented 77% of the coded emotional responses.

More than a decade later, Monray argues that the managerial lesson remains relevant: companies should not simply count complaints. They should understand what those complaints are saying.

Q: Why study customer complaints rather than customer satisfaction?

Monray: Because complaints contain an enormous amount of information. When a customer is satisfied, they may simply say, "Everything was fine." When a customer is dissatisfied, they often explain what happened, why they believe it happened and how the experience made them feel. That creates a much richer body of qualitative information. The challenge is how to analyze it systematically.

Q: What made the Capital One case particularly interesting?

Monray: The volume of available customer comments. We were able to examine 1,633 written reviews relating to Capital One credit cards over several years. That gave us a substantial textual database rather than a small number of individual interviews. The objective was to identify recurring patterns in the language customers were using.

Q: You used qualitative data analysis rather than simply counting keywords. Why?

Monray: Because words have to be interpreted in context. A customer may mention an interest rate, a fee or a customer-service representative, but the important question is what that issue represents to the customer. Is it frustration? Distrust? A feeling of being treated unfairly? We therefore used inductive coding. Rather than beginning with a completely predetermined list of categories, we allowed the themes to emerge from the material and then organized them systematically.

Q: And what emerged from the analysis?

Monray: We identified 38 codes associated with negative issues and six principal feelings. The emotional component was particularly interesting. Frustration, extreme frustration and anger were the largest category, representing 38% of the coded feelings. Another 24% expressed the perception that Capital One was fraudulent or had breached an agreement. And 15% expressed the feeling that the institution was hiding from customers or failing to confront problems properly. Those three categories represented 77% of the coded feelings.

Q: That's striking. What does it tell a manager?

Monray: It tells you that a complaint is rarely just about the technical issue that appears on the surface. A customer may complain about an interest rate or a charge. But underneath that complaint there may be a much more serious perception: I don't trust this company. That distinction is extremely important. If management fixes the immediate problem but does nothing about the underlying perception of unfairness or distrust, the customer relationship may remain damaged.

Q: So companies should analyze the emotion behind the complaint?

Monray: Exactly. Traditional complaint management tends to focus on the operational question: What went wrong? A more sophisticated approach also asks: How did the customer interpret what went wrong? Those are not necessarily the same thing. Two customers can experience the same operational problem and react very differently.

Q: What role does language play in your research?

Monray: Language is the raw material. Customers tell organizations what they think through words. The challenge is converting thousands of individual comments into structured information that managers can use. Qualitative analysis allows you to move from individual statements to recurring categories and relationships. In that sense, customer language becomes a form of organizational intelligence.

Q: Could this approach be applied beyond banking?

Monray: Absolutely. The methodology is not specific to Capital One. It could potentially be applied to airlines, telecommunications, hotels, universities, healthcare organizations or virtually any service business where customers generate written feedback. The industry changes, but the underlying managerial question remains the same: What are customers actually telling us?

Q: Today, companies have millions of online reviews and social-media comments. Does that make your approach more relevant?

Monray: It certainly makes the volume of available information much larger. But more data does not automatically mean better information. The challenge is still interpretation. Companies can collect thousands or millions of comments, but if they simply count positive and negative words, they may miss the meaning behind those comments. Modern text analytics and artificial intelligence can process much larger datasets, but the fundamental research problem remains: identifying meaningful patterns in human perceptions and emotions.

Q: Would you describe this as an early form of sentiment analysis?

Monray: There is certainly a connection. The research was concerned with identifying feelings and sentiments within naturally occurring customer text. At the time, the methodology was based heavily on qualitative analysis and coding. Today, technology allows researchers to process much larger datasets and automate parts of that process. But I would not say that technology eliminates the need for qualitative interpretation. It changes the scale of the analysis.

Q: What is the biggest mistake companies make when handling complaints?

Monray: Treating the complaint as an isolated transaction. A complaint is also a signal about the relationship between the customer and the organization. If a customer says, in effect, I believe you have treated me unfairly, the issue is no longer simply a billing problem. It is a relationship problem. That distinction should influence how management responds.

Q: What did this research ultimately teach you about consumers?

Monray: That consumers are not simply rational calculators. They interpret events. They assign motives to organizations. They develop perceptions of fairness, trust and credibility. And those perceptions influence their subsequent behavior. For a company, that means the technical resolution of a problem is only part of the solution. You also have to consider what the customer believes happened.

Q: If you had to give a CEO one recommendation based on this research, what would it be?

Monray: Don't ask only how many complaints you received. Ask what the complaints mean. Look for recurring issues, but also look for recurring emotions. If customers repeatedly use language associated with frustration, distrust or perceived unfairness, that is not merely a customer-service problem. It may be a strategic warning.

Q: And perhaps that is the larger lesson of the research?

Monray: Yes. A complaint is usually viewed as something a company wants to eliminate. I would look at it differently. A complaint is information that the customer is giving you for free. The managerial question is whether you are sophisticated enough to understand it.