Questions firms and their clients ask

A shared decision record for consulting and implementation projects. It reads your client call transcripts, suggests each decision with the exact words said, and counts a decision as agreed only when your firm and your client have both confirmed the same wording. Then it locks.

No. The client's reviewer confirms each decision, or asks a question, from a one-click email link. They can also sign in through their own company login if they prefer.

On test meetings it was never tuned on, the offline reader finds most decisions and actions, and it has never invented an agreed decision. Exact numbers, including where it falls short, are on the accuracy page. Every suggestion is checked by your project lead before the client sees it, so the record only contains what both sides confirmed.

Microsoft Teams, Zoom and Google Meet transcripts, Fireflies (automatically) and Otter or any notetaker export. You can also import an existing Excel or CSV decision log.

If Sealbook runs on your own server, no: we have no access. If we host it, the transcripts are on our server and someone with database access could read them today; we are adding encryption with a key only your firm holds. The security page explains both.

The rule-based reader runs entirely inside Sealbook. The optional Claude reader sends transcript text to Anthropic's API only if your administrator turns it on; under Anthropic's commercial terms, API inputs are not used to train models.

An agreed decision can't be edited. A later difference becomes a change request against the original, which both sides approve. The original stays visible, marked as replaced.

The six-week pilot is free. Pricing will be set with the first pilot firms, per active project.

Yes. It's one Docker image and a Postgres database, with sign-in through Microsoft Entra ID or any OpenID Connect provider. Setup takes about an hour.

Yes. Export the whole project, then delete it, or set an archived project to delete itself after a number of days.