The answer
“Working in dev” and “working in production” are different claims. Dev tells you the happy path runs. Production tells you what happens across thousands of real sessions, real users, and weeks of drift. Per-call logs can't answer the second one; they show that calls succeeded, not whether behavior is healthy.
Flowlines starts with four operating metrics: analyzed chats, active users, issue-free rate, and negative feedback. It then shows user cohorts, intents, issues, and the sessions that need attention.
Open an issue to see what happened, why it matters, the next step, where it fires, and the original sessions behind the pattern.
What should I check every week?
- The share of sessions with a completed outcome, and how it moved.
- Unresolved sessions and the users they affected.
- The top recurring issues, ranked by how many users they reached.
- New behavior patterns that appeared after the latest release.
- Cost per completed task, so quality gains are not paid for twice.
Published