Purchases made via links on this site may earn us commission.

Providing support to employees can be a cumbersome venture for a company. It is particularly complex for a company like T-Mobile. There are phone features, plan offerings, and promotional deals that change near daily. Properly securing systems that handle sensitive customer data (something T-Mobile still needs a lesson or two about) is also important. There is even financing math to get right.

It is a lot to remember and get right across retail and care channels, and now T-Mobile has a plan to make this a more accessible experience for employees.

Enter frontline “Superpowers”, new digital tools to make the mountain of necessary information more manageable. In a few internal documents provided to us here at The Mobile Report, we have details of three new digital tools rolling out to employees.

GenAI Chat

  • Enhances the customer experience by quickly providing answers to complex questions
  • When asked, provides verbiage directly from policies to help you through difficult interactions
  • Analyzes content across The HUB (a sort of “front page” for employees), C2 (the main internal documentation service), and more to ensure you’re the T-Mobile Superhero!
  • Available in Retail and Care

Promo Genius

  • Streamlines the decision-making process to allow for quick, efficient promo matching tailored to individual customer needs
  • Dynamic filters optimize promo results, empowering you to swiftly pinpoint the most relevant promos, all while emphasizing value
  • Gives you the power to right-fit promotions more efficiently so you can create the best experience for your customer
  • Available in Retail and Care

Next Best Action (NBA) in Retail

  • Recieve real-time recommendations based on hundreds of data points
  • Increase Customer Love with a personalized experience
  • Forget what you’ve used in the past – this is a brand-new tool for Retail, serving up way more than just promotions
  • The more you use NBA, the better the recommendations will be
  • Experience Stores Only: Use NBA in Tapestry and not NBA in Atlas
  • VR (Virtual Retail) channel not in scope.

An education in AI prompt writing

A big caveat to making AI chatbots work is how you tailor your questions. Beyond training on general access and use of the tool, T-Mobile is also training employees in basic ‘prompt engineering’ – the act of conversing properly with AI to maximize the information it can provide. Communications include charts on how to improve basic AI prompts to glean better information from the tool.

Prompt engineering training for T-Mobile frontline

Employees are encouraged to use GenAI Chat for some common customer interactions.

  • You need an answer to help solve a customer issue.
  • Get help on a product or service.
  • Discover how top Experts communicate with customers.
  • Compare devices or rate plans.
  • Whenever you need to find information about anything T-Mobile!

T-Mobile’s backend automation that feeds data to the AI is also collecting interaction data from other customers as a reference point to give advice to anyone using the tool. How else could it respond with how other employees are communicating with customers?

While the value is certainly there as a training and optimization tool, the hope is that customer-proprietary information cannot slip through. It is powerful, but prompt hacking has become an increased discussion with the rise of AI. Does a truly safe LLM exist yet? Especially in an industry rife with cyber attacks?

Employees aren’t too sure about their new AI overlords

In a now-deleted Reddit thread, an employee posted about the chatbot asking other employees for feedback. Here are some of the top comments.

It sucks. You have to be very specific with your questions and more often than not it will quote with outdated information and promos instead of what’s currently available.

I’m sticking with just using the upgraded search for c2.

Reddit user

They’re pushing leaders to use AI for documentation and observations and honestly, if as a leader you need an AI to objectively give your team feedback, you probably shouldn’t be a leader in the first place.

Reddit user

The only thing that seems useful is the specification comparison. I get tired of customers asking what the difference between the IPhone 13 and 15 is, so now I can just show them

Reddit user

The initial sentiment appears rough for those vocal enough to leave a comment.

T-Mobile Isn’t Waiting For Change At Least

These retail “Superpowers” appear to be a move that plays into a bigger automation strategy for T-Mobile’s future. Additional screenshots shared with us show even more backend automation for other initiatives.

Other T-Mobile automation initiatives

Expert Head Start

Expert Head Start uses Network, Billing, and Customer data to predict the reason for a customer contact before we even say hello, and to assign a next best action for that customer.

Expert Recap

Expert Recap uses generative AI to summarize an entire call for Experts, leaders, and customer to reference later, including a summary of what was discussed, actions taken, and what the customer can expect.

Only time will tell if T-Mobile’s attempt to embrace digital tools to help their employees will end up being worth the investment. There is a silver lining through this period of transition, though.

In the future, ensuring you get your promotions matched up right the first time could be much smoother. Mobile Experts in store or Support Representatives can get answers without lengthy excursions to escalation teams. Even if you get an outsourced agent, this could lead to a more consistent experience.

In the long run, this is the gamble T-Mobile has taken to unlock the potential of AI as a support-assistant tool for frontline retail. Let’s just hope that all this training isn’t leading to more layoffs when the AI learns enough from mass use. While not a labor cost-saving measure yet, who knows what comes in time. GenAI Chat is hopefully not a first step toward more challenging labor optimizations.

Is T-Mobile on the right track bringing GenAI as a frontline support tool? Are these other automation initiatives going to see traction for them? Do you want AI to be providing your support? Let us know in the comments below.

Enjoy this post? Consider supporting us on Patreon!
Become a patron at Patreon!