The coordination cost of managing tech debt manually at global scale
BHN's platform team operates in one of the most demanding environments in payments: high transaction volume, strict compliance requirements, and a multi-tenant estate that is constantly evolving. The team had the right people, the right tools, and disciplined operational controls in place. The challenge was not capability. It was that even a well-run tech debt practice creates more friction as it scales when it depends on periodic effort and manual coordination.
Three forms of that friction were becoming increasingly visible:
Maintaining an up-to-date view required constant manual effort
The team carefully tracked service versions, dependencies, infrastructure components, and lifecycle states. But doing that through scripts, queries, and cross-team coordination produced accurate point-in-time snapshots rather than a continuously current system of record. As the stack evolved with new services, integrations, and cloud resources, maintaining that view meant repeating the work rather than simply reading it from the system.
Determining ownership was a recurring task, not an instant answer
When a risk surfaced, a library approached end of life, or configuration debt appeared, the team knew how to route it. But mapping each item to its current owner across fast-moving platform and product teams still had to be reconstructed each time. This was something the team wanted resolved automatically.
External triggers shortened planning horizons
Much of the team's tech debt activity clustered around external triggers such as vendor end of-support dates, security disclosures, and audits. Even when handled well, those triggers compressed timelines and pulled senior platform engineers toward coordination and firefighting instead of FinOps, developer experience, and broader platform capability work.
