Observability belongs in the product, not a status page

When a payment is pending, finance asks engineering why, engineering opens a separate observability tool, and the two teams reconcile screenshots. We deleted that gap by putting request logs, latency, and delivery status inside the same console both teams already use.

The change was cultural as much as technical. When everyone looks at the same request log, "why is this pending?" becomes a shared question with a shared answer, not a support ticket.