How AI can help reduce customer churn at call centres

Keatext's sentiment analysis helps call centres maintain healthy retention, reduce customer churn, and better understand customer behavior.

reduce customer churn

Illustration by Isaac Claramunt

When it comes to customer churn, call centres often function as the proverbial front lines in the fight to maintain a healthy retention rate, making it crucial to equip agents with the tools they need to best respond to what can be a business’s most volatile customers.

The disruption of the traditional call centre model is already underway, with chatbots, automated messaging and voice response sweeping up the easy-to-answer low-hanging fruit of the customer care world—your FAQs and your quick fixes, if you will—and fundamentally shifting the foundation on which the call centre model was built.

But like all work environments molded and chiselled by the advancement of automation and artificial intelligence, the effects aren’t all grim for the real people who work in the industry. 

Sentiment and context are some of the hardest things to determine

The implementation of AI text analytics technology tends to have the counterbalancing effect of making the work of employees all the more impactful and all the more important, helping them not just provide answers to customer qualms but effectively reduce the churn rate of a business.

With automated services picking up the customer service slack with simple solutions to less consequential problems, agents’ time can be freed up to effectively manage more pressing pain points. But technology is just as critical at this stage of the customer service process, where companies have potentially the greatest opportunity to keep customers happy and retain their business.

Why churn matters

Anita Toth, a consultant specializing in helping businesses reduce customer churn, uses the metaphor of a relationship to get to the heart of how customer service can play into churn.

“When you’re dating somebody, there’s a lot of attention paid to you and then once you get into the relationship, some of that tension tends to dwindle off,” she says. “It’s really similar here: once people become customers, they tend to not get as much attention paid to them, there are gaps in how often they communicate, and they may feel they’re not as valued.”

What happens next represents, in Toth’s view, both the greatest risk and the greatest opportunity for businesses regarding their customer base: effective communication can make all the difference between keeping a customer happy or handing them off to your competitors.

Call centre agents can identify the root cause of an issue, relying on customer feedback analysis tools designed to do just that.

“Sentiment and context are some of the hardest things to determine,” explains Toth. “The best way to reduce churn is to ask more probing questions to get an idea of what it is that is actually bothering the customer.” Existing technology around sentiment analysis has the potential to not only better position call centre agents to manage unhappy customers, but can also help businesses maintain a better handle on customer satisfaction and potential sticking points. “66% of customers are happy to give a business a second chance if they address their real underlying issue,” says Toth. “But you have to dig for it.”

How data can help

66% of customers are happy to give a business a second chance if they address their real underlying issue, but you have to dig for it.

The first step, according to Toth, lies in qualitative data collection—for which call centres can be a veritable gold mine. “There has to be a system in place where calls can be categorized broadly,” she explains. “This is where you can use AI to figure out what the main themes are. You are going to notice patterns that come up, that people tend to be complaining about or struggling with, and you can start categorizing them.”

Toth describes a scenario where a business might have ten buckets under which complaints can fall into, which establishes context within which agents can operate effectively, answer complaints intelligently, and ultimately, help customers feel heard. This way, call centre agents can more efficiently identify what could be the root cause of an issue, relying on customer feedback analysis tools designed to do just that. “Once you get those categories in place, you can start developing a system so that when a support call comes in, the agent can look at the broad themes and ask probing questions to come to a resolution much faster.”

Toth also cautions her clients not to underestimate the true cost of customer churn. From Toth’s perspective, 80% of customer churn can be reduced, and much of it comes down to effective communication. With call centre agents armed with the tools they need to effectively listen, capture customer data and understand pain points, plus a strategy to maintain communication before things go sour, businesses have ample opportunities to keep the customers they worked so hard to court in the first place.

“It’s about making a conscious effort to keep in touch with customers,” says Toth. “The data then becomes so invaluable because you can start seeing patterns in what people are saying, and catching things before they become a problem.”

Keatext decorative accent

Make your contact center more profitable with Keatext

Explore ticket analysis with Keatext

Related Stories

Keatext decorative accent

Subscribe to our blog