Skip to main content
After a conversation ends, the platform runs an LLM pass over the stored transcript and produces three results: a summary, values for the data fields you define, and a verdict for each success criterion. Summaries are on by default for every agent — extraction and evaluation are opt-in, configured per agent on the Builder’s Analysis page.

See it on a test call

End a preview call and watch the analysis card fill in.

Read results via API

Analysis is embedded in every session detail response.

Push to your backend

Get notified the moment a conversation is analyzed.

What a run produces

Summary

A concise recap of the conversation. Enabled by default — every conversation gets one out of the box.
  • Custom prompt (optional, up to 1,000 characters) — replace the platform’s default summary instruction with your own, e.g. “Summarize in two sentences, focusing on the caller’s request and the outcome.”
  • Output language — fixed per agent, chosen from 10 locales (English, Chinese, Japanese, Korean, Spanish, French, German, Portuguese, Russian, Arabic). Default is English.
The summary language does not follow the conversation language. Summaries and rationales are always written in the configured language, even when the conversation happened in another one — so downstream systems see consistent output.

Data fields

Structured values extracted from the transcript — up to 20 fields per agent. Each field is: When the conversation doesn’t contain the information, the field’s value is null with a rationale explaining why — the model never guesses.

Success criteria

Up to 10 criteria per agent, each a name plus a description (up to 500 characters) stating what a successful conversation looks like — “The agent resolved the caller’s issue or set clear next steps.” Every criterion gets a three-state verdict with a rationale:
  • success — the transcript shows the expectation was met.
  • failure — the transcript shows it was not.
  • unknown — the transcript is incomplete, the answer was ambiguous, or the information needed to judge is missing.
The unknown state is deliberate: rather than forcing an ambiguous call into a binary verdict, the model tells you it couldn’t judge — so your success-rate numbers stay honest.

Configure analysis

Open your agent’s Analysis page in the Builder. Three cards — Summary, Data fields, and Criteria — autosave to the draft as you edit. You can also set the analysis section of the configuration via the API:
Like all configuration, edits land in the draft. Preview calls use the draft immediately; production sessions pick up analysis changes after you publish. To disable analysis entirely, turn the summary off and leave both lists empty — conversations then finish with analysis status skipped and no LLM call is made.

When analysis runs

Analysis starts automatically when a conversation ends — production sessions from any source, and preview calls when you end them. A run moves through these statuses: Results are typically ready within seconds of the conversation ending.
Production sessions are analyzed against the agent version pinned when the session started — editing your criteria afterwards never changes how past conversations were judged, and each result stays attributable to the configuration that produced it.

Read the results

In the Builder, analysis appears in the call panel after a preview call ends. Programmatically, the session detail response embeds it under analysis:
To push results into your CRM or data warehouse as soon as they’re ready, configure a post-call webhook — it fires once analysis completes.

Going further

Conversation history

Retrieve stored transcripts and session details over REST.

Webhooks

Deliver analysis results to your systems automatically.

Preview calls

Iterate on fields and criteria with instant feedback after each test call.

Versions & publishing

How draft edits roll out to production sessions.