Scattered customer issues pointed to upstream operational gaps. I connected the signals, clarified routing and ownership, and helped reduce incoming tickets by 15%.
In AWS Billing, customer issues surfaced through scattered channels - testing, support, ticket queues, system events, partner teams - and each one looked isolated. Payments Operations also wasn't always included early in launches that affected payments, so issues arrived later as customer contacts and tickets.
Incoming tickets dropped 15%. The broader 25% target wasn't reached - a single launch or system event could swing volume - which reinforced the real lesson: connect signals upstream, don't just manage them after they become tickets.
Customer issues rarely begin as escalations. The work is connecting testing, operational data, and customer feedback early enough to hear what they're telling you.