CASE STUDY 01 · T-MOBILE · ENTERPRISE TELECOM
T-Mobile's enterprise platform had been built system-out, not user-in. Admins had to know which backend or portal owned a service before they could manage it. We rebuilt the experience around what admins were trying to get done.
faster for admins to get up and running
fewer logins into external systems
services managed from one workspace
01 · CONTEXT & CONSTRAINTS
Enterprise customers bought connectivity, security and 5G services from T-Mobile, but each service lived in its own internal system with its own logic. T-Mobile had no control over how these systems worked or looked, and no visibility of the problems customers hit. Frustrated enterprise customers were starting to look at other providers.
The job wasn’t a new portal with a fresh UI. It was an experience layer that hid backend fragmentation without touching the backends.
fragmented telecom services
internal systems behind them
billing accounts in a large enterprise
distinct admin roles
WHAT COULD NOT CHANGE
02 · THE PROBLEM
Admins had to know which system owned a service before they could do anything with it. The platform mirrored the org chart, not the job.
USER PAIN
“We know there is an issue with service but don’t know what is the issue and when will it get resolved”
BUSINESS COST
The average task meant switching between at least two systems.
HOW WE KNEW
An audit of 20+ services and their portals, plus 2 stakeholder workshops.
THE FINDING THAT CHANGED DIRECTION
The original brief was a unified portal with a UI refresh. Early research showed the existing platforms were system-centric, not user-centric, so a reskin would have preserved every broken path. We reframed the brief around a user-centric unified portal, built on real user journeys rather than system structure, and T-Mobile team agreed to the shift.
03 · KEY DECISIONS
D1
Navigation mirrored 15+ backends. We grouped every service under user intents such as Connectivity and Security so admins could act without knowing the system behind it.
TRADE-OFF
Scalability vs. simplicity. The accordion-style navigation scaled well for large portfolios, but service owners with fewer than five services might find it a little complex.
D2
Large enterprises held thousands of billing accounts, and flat tables broke down. We added a global context switcher with layered filters, so admins could drill from a corporate view to one account without losing their place.
TRADE-OFF
Speed vs. orientation. Jumping from company to account in two clicks made it easy to lose track of where you were, so the selected account and its organisation stay pinned in the header.
PROOF
35% faster for admins to get up and running
D3
With 36 services built in parallel by different teams, shared buttons were not enough. We defined patterns for charts, tables, bulk editing and service provisioning so behaviour stayed consistent whichever team member built the feature.
TRADE-OFF
Consistency vs. rework. Keeping patterns consistent meant repeatedly revisiting services we’d already released.
PROOF
Design-related bugs fell by 50%, even across multiple products.
04 · THE SOLUTION
05 · OUTCOME
faster for admins to get up and running
fewer external-system logins
faster customer onboarding
increase in cross-service adoption
“Let’s redesign Account Hub exactly like this.”
After a year of positive feedback, T-Mobile decided to merge its main portal, Account Hub, into the new workspace.
06 · WHAT I'D DO DIFFERENTLY
We assumed admins needed power features like column customisation and saved filters for complex, data-heavy tables. Usage data showed most relied on simple filters to get the job done.
We built dashboard layout editing and theme selection without research. Next time, I’d validate demand first and weigh the impact against the cost of building it.
The team is now adding more services, along with service-agnostic features that work across the platform.