Operational remediation
Resolving 12,000 legacy account exceptions—safely.
A long-running workaround had created thousands of unusual account conditions. The real challenge was deciding what could change without creating new customer or financial harm.
The challenge
Not every exception meant the same thing.
Some lines still had service. Some carried retail installment contracts. Others had remained on temporary device identifiers for years. A blanket cleanup could have triggered immediate charges or disrupted legitimate customer arrangements.
The issue required more than data correction. It required a way to distinguish risk, decide appropriate action, communicate with customers, and protect against system behavior that could recreate the problem.
The approach
Turn ambiguity into decision logic.
I defined the reporting requirements, worked with the reporting team to produce the necessary data, and established the cadence for reviewing it.
I then translated the account conditions into actionable groups—what could close, what required protection, and what needed further review—while coordinating finance, legal, communications, fraud, analytics, and operational stakeholders.
- Reporting requirements
- Review cadence
- Action criteria and decision logic
- Contract and customer safeguards
- Automation-behavior correction
- Customer communication paths
- Recurring processing
- Documented ownership and handoff
The principle
Cleanup should not create a new mess.
The defining choice was restraint. Accounts with active contracts could not be handled like dormant, contract-free lines. Automation that restarted billing after suspension also had to be addressed so it did not work against the resolution process.
By combining data, operating rules, safeguards, communications, and ownership, a one-time problem became a repeatable process another team could sustain.