Using Technology to Improve Personal Connections
T-Mobile prides itself on being a disruptor in the world of wireless communications, always thinking creatively about the relationship it wants to have with its consumers. That includes the company’s approach to using AI for customer service.
Using the predictive capabilities of machine learning to improve customer service is a great example of AI augmenting human abilities. T-Mobile sees it as an opportunity to serve customers better and faster, benefiting not just the company and its service agents but also enriching the customer experience and creating stronger human-to-human connections.
“Most industries have looked to use AI and machine learning to build more sophisticated Interactive Voice Response (IVR) systems and chatbots as a means to deflect for as long as possible the interaction between a human customer service agent and the customer,” says Cody Sanford, executive vice president and chief information officer at T-Mobile. But at T-Mobile, they flipped that paradigm. T-Mobile customers immediately connect with a customer service agent that knows them, rather than talking to an IVR or chatbot. With the help of AI, these customer service agents can quickly access the information most salient to customer needs.
Providing agents with contextual information in real time helps guarantee that each customer’s issues are quickly and accurately resolved. To do this, T-Mobile developed Natural Language Understanding machine learning models that extract meaning from vast amounts of textual data. The company’s data includes hundreds of thousands of incoming customer requests a day, as well as knowledge repositories where potential answers to customer queries can be found. The machine learning models then predict which information would serve a specific customer’s needs, such as helping them pay their bill or adding a new phone line, and then the relevant content will be displayed to the customer service agent who is part of the tight-knit Team of Experts (TEX) group that knows the customer because they regularly work with them.
"T-Mobile’s customers like it when they have a personal, human connection with us. Through machine learning, we can reshape how our customers relate to us."
Cody Sanford
EVP and CIO
T-Mobile
"T-Mobile’s customers like it when they have a personal, human connection with us. Through machine learning, we can reshape how our customers relate to us."
Cody Sanford
EVP and CIO
T-Mobile
However, before this process can begin, labels need to be added to the data in order to train these predictive machine learning models. Previously, T-Mobile had teams of data scientists working on manual labeling. It was vital work, but also tedious and time-consuming. The data scientists combed through customer messages looking for key words and phrases and mapped them to transaction types.
To infuse its data labeling with AI, T-Mobile turned to Amazon SageMaker Ground Truth. Ground Truth speeds up and scales labeling of training data, which is essential for machine learning models to produce predictions with high accuracy. Instead of doing this manually, Ground Truth learns from these annotations in real time and automatically applies labels to much of the remaining dataset.
Using Ground Truth not only streamlined that process, but freed T-Mobile’s data scientists to focus on more specialized tasks, such as model creation, analysis, validation and deployment.
“We were spending copious amounts of our data scientists’ time labeling thousands upon thousands of messages,” Sanford says. “Ground Truth has been able to make that super-efficient, and now we don’t need our skilled data scientists hand-labeling data anymore.”
For example, Ground Truth creates accurate training data by looking at phrases and keywords contained in millions of customer text messages, helping T-Mobile build better predictive recommendations about why a customer is reaching out, so that the right answer can be delivered on first contact. The model is designed to be self-learning, so it will get more and more accurate over time.
“T-Mobile’s customers like it when they have a personal, human connection with us,” says Sanford. “Through machine learning, we can reshape how our customers relate to us.”