Claap already gives you the transcript, the summary and AI answers on every call. But the questions that actually move a deal are yours: did this call clear the budget bar? What objection came up? Is there a real next step? A generic summary won't answer those the same way across 200 calls — and turning raw AI loose on transcripts gets you an answer that sounds right and is often wrong.
🤖 Build any AI field, on any view
AI fields — the evolution of Insights — let you add an AI-generated column to any meeting or deal view. Write a prompt once, and Claap runs it against every recording in the view and returns a structured value: a rating, a status, a short text, whatever you need.
What you can do:
- Start from a template or from scratch. Pick a ready-made field or write your own, then set its icon, name, type and prompt.
- Score on your framework, not a black box. Build MEDDIC, SPICED or your own scorecard as fields — rating + gap + proof — and see exactly why each call got its score.
- Extract the signal you care about. Objection, next step, champion, competitor mentioned, sentiment — one field per signal, across the whole view.
✨ Test it, generate it, auto-run it
What's new:
- Test before you run. Preview a field's output on a few records and tune the prompt until it's right.
- Generate across the view. Run a field on every row, or only the empty ones, in one click.
- Auto-run on new calls. Turn it on and every new recording in the view gets the field filled automatically.
- Share across your workspace. Publish a field so the whole team scores calls the same way.
📊 Sync to your CRM, and pull the whole view into your AI
Map any AI field to a HubSpot or Salesforce property, and let auto-run keep your CRM updated after every call — no manual admin. And because your fields are structured and sourced, you can pull the entire enriched view through Claap's MCP into Claude or any agent.
That's the difference between a plausible answer and a precise one. Instead of turning raw transcripts loose on your AI — where the error risk multiplies with every call — your agent reasons over clean, sourced fields (rating, gap, proof), so cross-call analysis actually holds up.
💡 Pro tip — build one "deal risk" field with a rating and a one-line reason, turn on auto-run, then ask Claude to list every deal rated high-risk this week. Grounded, in seconds.
AI fields are live today on Pro and Business. Open any meeting or deal view, add a column, and pick AI field to start.
