Case Study: Scribe cuts debugging time to 5 minutes and observability costs 75% with Honeycomb
Key results
The challenge
Scribe, an AI-powered documentation and process-automation platform, ran a rigid observability stack where troubleshooting during incidents could take 30 to 60 minutes. Observability was treated as something to add later rather than built into engineering workflows.
The solution
Scribe adopted Honeycomb for end-to-end distributed tracing, letting engineers follow a user's full workflow and see exactly where in the system a request was slowing down.
“When we have an incident, Honeycomb is our trusted source of truth. It's not even a question anymore—Honeycomb is where our investigations start.”
STSteven TanSenior DevOps Engineer, Scribe
The results, in context
Scribe reported that the average time to identify root cause during incidents fell from about an hour to five minutes and that observability costs dropped 75%. Roughly 90% of engineering adopted the full feature set within months.