T-Mobile, improving customer experience with AI and IT Operations
Together with T-Mobile, Ericsson uses AI to minimize order fallouts in multi-vendor business and operations support systems (OSS/BSS) and to find and resolve issues faster to improve the customer experience.
Reducing order fallout using AI
T-Mobile was in a challenging situation with regards to their ordering process. Up to 17% of all orders were falling out, anything from upgrading a price plan to getting a new phone, which had a detrimental effect on the customer experience. In a large and complex system landscape with over 140 multi-vendor applications, ranging from CRM, Order Management to Activation, it was hard to pinpoint the issues using traditional methods. Together with T-Mobile, Ericsson took on the task to move beyond old-school thinking, combining cutting edge telecom and IT operations skills with AI and machine learning to solve the problem.
Get the story behind the project with Jason Lichtenberger, Manager Product & Technology at T-Mobile, Jaime Burgoa, Managed Services Director at Ericsson, and Ayman Abolfadl, Vice President & Head of ICT T-Mobile US at Ericsson.
The results from the project were outstanding. Together we could reduce: order fallouts by 95%, time to identify an issue by 90% and the order-to-activation process was shortened to 5 minutes or less for almost every order. T-Mobile now has a long-term solution, giving a clear end-to-end view of the order flow and the ability to swiftly pinpoint where any new potential issues could arise. Together, we delivered on T-Mobile’s promise of being the uncarrier, providing the best possible customer experience to subscribers.
- Jason Lichtenberger, T-Mobile
An AI-powered solution across the whole E2E order flow
Order fulfillment is a complex process that requires holistic thinking, working with the entire order flow and across all applications to create a long-term, sustainable solution.
“The main idea was not to resolve issues as they were coming up, but to find a scalable long-term, proactive, anomaly detection solution, built on a solid foundation of AI”, says Jaime Burgoa, Ericsson Managed Services.
The project was divided into three key phases - Assess, Solve, and Improve.
The Assess phase consisted of a human-led study aimed at creating a heat-map of where issues emerge to understand where to deploy our AI tools.
In the Solve phase the AI algorithms found patterns in the data using anomaly detection. With these patterns we could predict where problems would arise and preemptively resolve them using closed-loop automation.
In the Improve phase, the Ericsson and T-Mobile teams leveraged the deployed AI & Automation to improve customer experience by reducing order fallout, improving mean time to identification and improving activation time.
- Jason Lichtenberger, T-Mobile
