Reducing customer churn using text analytics

In our ebook What Method? What Tools? A Guide To Analyzing Your Customer Feedback we introduce Lucy, the head of Customer Support at a medium-sized telecom service provider. Lucy has been tasked with investigating why, lately, customers are leaving and switching to another provider. In order to find the solution to reducing customer churn, Lucy knows there’s a wealth of information buried in her customer support data but with 250­ 000 tickets collected over 18 months—that’s nearly 14 000 tickets a month—reading through each piece of feedback seems infeasible (and rightly so). At the end of our ebook, we leave it up to the reader to decide what solution Lucy goes for, but let’s assume she goes with Keatext. (This post would be very short otherwise.)

Reducing customer churn: A job for Keatext

The question is now: How does Lucy use Keatext to reduce customer churn?

Well, Lucy probably starts off by signing up for Keatext and is faced with yet another decision: what plan should she choose? With 14 000-something tickets every month, and with a growing business, she is likely to exceed our low-range plan’s monthly record quota, and she would save herself a lot of trouble by picking our mid-range plan, which allows her to analyze up to 50 000 tickets a month and use any one of our CRM integrations, as well as analyze all of her historical data for a one-time fee. Mid-range plan it is!

After signing up and logging into the app, Lucy starts fetching data from her favorite CRM through one of our connector apps. Keatext is a fast reader, but it still needs a few minutes to process Lucy’s 250 000 tickets. While Keatext compiles the results, Lucy can start creating topic groups and blacklists to reflect her organizational knowledge—results will be neater and more relevant to Lucy’s interests.

reducing customer churn

Once Keatext is finished, Lucy dives in. In an effort to find a solution for reducing customer churn, she wants to find out why customer satisfaction is so low, and she thinks it might be because of the way support agents interact with customers. Earlier she told Keatext to treat topics like employeerepresentative and agent as synonyms, and now she can see that her staff is mentioned negatively 73% of the time. She clicks on staff and finds out that customers find the staff pushy, uninformed and sometimes difficult to understand.

reducing customer churn 2

reducing customer churn 3

Can we pinpoint these issues to a specific call center? To find out, she filters results based on location and sees that people complain the most about the Canadian branch. To wrap up, she reads a few tickets to find specific complaints about her staff, then she exports all relevant tickets to show her colleagues and come up with ways to improve customer support experience for Canadian customers.

Join Lucy and the hundreds of people who use Keatext’s powerful text analytics features to make sense of their customer feedback. If you haven’t already, give our ebook a read, then head over to the application for a free 14-day trial to see if Keatext is right for you!

 

FRESH CONTENT DELIVERED STRAIGHT TO YOUR INBOX

0.Comments

    Leave a Comment