Executive · Follow-ups

Every commitment from the meeting, owned and dated

The decisions were good. The problem is the eleven commitments that existed only in a document nobody reopened.

What the worker does

Meeting follow-ups in Polaris start from notes in Docs. An AI worker with the Google Calendar connection reads the notes, proposes one task per commitment with an owner and a due date, and delivers them as prefilled suggestions in the Inbox for one-click acceptance. It also reports commitments from earlier meetings that never moved, which is the part everyone avoids.

Connections
Google Calendar · Slack
Output
One proposed task per commitment
Also reports
Older commitments with no movement

Notes are not follow-up

A good note-taker produces a document. A document is a record, not a mechanism. Nothing in it has an owner, a date, or a place on anybody's board, so the commitments depend entirely on each person remembering their own.

The conversion step, reading the notes and creating eleven tasks with the right owners, takes twenty minutes and is nobody's job. So it happens after important meetings and not after ordinary ones, which is where most commitments live.

From notes to a board

  1. 1

    Notes land in Docs

    Written by whoever was in the room, in the nested doc tree next to the workstream the meeting belongs to.

  2. 2

    The task fires after the meeting

    The worker reads the notes and the calendar entry, which tells it who was present and therefore who can own something.

  3. 3

    Commitments become suggestions

    One prefilled task per commitment, with bucket, lane, owner and a due date already set. Anything the notes left ambiguous is flagged as needing an owner rather than assigned to a guess.

  4. 4

    One click each

    Whoever ran the meeting accepts, corrects or rejects. Ten seconds a task, and nothing entered the board unapproved.

  5. 5

    The next report includes the ghosts

    Commitments accepted three meetings ago with no activity since are listed by name, which is the mechanism that makes the whole loop work.

Two versions of the same meeting

Notes only

  • Eleven commitments in a document
  • Owners implied by whoever was talking
  • No dates
  • Rediscovered when the same topic comes back in six weeks

Notes plus a follow-up task

  • Eleven proposed tasks, each with an owner
  • Dates that appear in the Focus lane
  • Ambiguous commitments flagged rather than assigned
  • Old commitments that never moved, named out loud

Questions people ask

+Does Polaris transcribe the meeting?

No. It works from notes in Docs and the calendar entry. Record the meeting however you already do, and put the notes where the worker can read them.

+What if the notes are vague about who owns something?

The worker flags it as needing an owner instead of assigning it to whoever spoke last. An incorrectly assigned task is worse than an unassigned one, because it looks handled.

+Can this run for every meeting?

It can, and the cost tracks the volume rather than a seat count. Start with the recurring meetings that generate commitments, because those are where things quietly disappear.

+Who sees the list of commitments that never moved?

It is a comment on a task in the workstream, so everyone in that workstream sees it. Making it visible rather than private is what stops the same commitment being re-agreed three times.

Your next hire takes 60 seconds.

The software is free — unlimited people, tasks, workstreams and docs. You pay only for work an AI worker actually delivers, itemised by the hour.

Get started free

Last checked .