Skip to content
Log in
Metrics & Measurement

How to Measure Whether Fin Is Actually Resolving Conversations

Share

Measure Fin performance with four metrics: resolution rate, Fin Surveyed CSAT, handoff quality, and repeat-contact rate within a window your team defines. Intercom's Holistic overview compares Fin and teammates side by side. Keep separate baselines when they handle different intents, because a blended score hides where the experience changed.

Four metrics for Fin performance

Metric Definition Where to get it
Fin resolution rate Fin-closed / Fin-assigned, no reopen in 72h Reports + reopen tags
Fin CSAT CSAT on conversations Fin handled Reports → Surveyed CSAT
Handoff quality % handoffs where human resolves without customer repeat Manual sample or IQS attribute
Repeat contact rate Same customer, same topic within 72h Tag or custom report

How to filter Fin in Intercom Reports

  1. Open the Holistic overview report for side-by-side Fin and teammate performance.
  2. Open Surveyed CSAT for response rate, rating volume, and satisfaction trends.
  3. Export weekly CSV; store Fin and human rows separately.
  4. Never compare Fin FRT to human FRT without noting Fin is instant on first reply.

Reporting gaps: Intercom reporting limitations. CSAT report walkthrough: build a CSAT report in Intercom.

Blended CSAT hides Fin regressions. Supportman tags Fin vs human in weekly Slack reports and scores unrated Fin threads with predicted CSAT.

Use the five-check Fin QA framework

Fin CSAT vs human CSAT

Fin CSAT often runs lower on complex threads because customers rate the outcome, not the bot's politeness. Set Fin targets from your own baseline after 30 days, not from human CSAT benchmarks.

Benchmark framework: Intercom CSAT benchmarks (human vs AI section).

When Fin looks good but CSAT falls

High Fin resolution rate with falling blended CSAT usually means Fin is closing threads that should escalate. Audit with how to QA Intercom Fin and check repeat-contact rate on Fin-closed conversations.

Fin CSAT FAQ

How do you measure Fin CSAT in Intercom?

Use the Holistic overview to compare Fin and teammate performance, then use Surveyed CSAT for rating volume, response rate, and satisfaction trends. Keep the two baselines separate when their intent mix differs.

What is a good CSAT score for Intercom Fin?

Benchmark against your own Fin baseline after 30 days. Fin often scores below human agents on complex intents; improving handoff quality matters more than matching human CSAT.

Does Fin affect team CSAT reports?

Yes, unless you filter Fin out. Blended reports make human coaching decisions unreliable once Fin volume grows.

Can you score unrated Fin conversations?

Yes. Intercom CX Score evaluates conversations without relying on surveys, and Monitors with Custom Scorecards can run QA against defined criteria. External tools can provide another predicted-CSAT or IQS model. See validate predicted CSAT.

Under two minutes to live, no IT ticket required.

See pricing
Prefer us on Google