The best applications of machine learning occur when algorithms augment human activities instead of replacing them. That’s the case with T-Mobile, which has deployed machine learning to help contact center agents better serve customers.
In this episode of the Ahead of the Pack podcast, Heather Nolis, Senior Software Engineer with T-Mobile, talks with Tim Crawford about the telecommunications company’s approach to artificial intelligence and machine learning and the impact they’re seeing across the business.
Nolis described one business problem engineers were tasked with solving: “If somebody has a problem on their network or with their phone and they call T-Mobile, how could we help those experts sitting in the call center solve your problem as quickly as possible, using all of the information that our company has at its disposal?”
Nolis and her team built machine learning models that sit between a care agent and the customer, scanning previous customer interactions and serving up relevant information to the care agent to help them quickly address the customer’s issue.
One interesting aspect of this deployment is that success metrics for the technology are the same as for human agents, such as first-call resolution.
“We can measure based on just the accuracy of the models, but that doesn’t actually help the business,” she added. “So the things we are held accountable for are the same KPIs our care agents are held accountable for: how satisfied was the customer at the end of the call? Did they have to call back again to solve their problem? Because [improving] the way that our customers feel about our business and the way they experience our business is what we should be aiming for all the time.”
Sponsored Links