Meetings & Communication · Updated July 17, 2026
ChatGPT prompt to decide if a meeting is worth having
This is the prompt we run before scheduling, or accepting, a recurring or long meeting. Its core rule: it assesses the meeting honestly, prices its fully-loaded cost, and returns a clear verdict, without assuming it should be cancelled.
The prompt to copy
Assess this meeting honestly. Purpose: {purpose}. Attendees: {names/roles}. Length: {minutes} min. Recurrence: {one-off / weekly / etc}.
1. What decision or output would justify it?
2. Could it be an async doc or message? If yes, draft that async version.
3. If it should happen: the minimum attendee list and shortest workable length.
4. Estimated fully-loaded cost: attendees × length at {average hourly cost — if I gave none, use $75/h and say so}.
5. Verdict: keep / shorten / make async / cancel — one line why.
Replace the fields in curly braces (purpose, attendees, length, recurrence), then run it.
When to use it
Before adding a new recurring meeting to the calendar, or when an inherited weekly slot hums along without anyone quite knowing what it is for anymore. Also useful before accepting a long invite: the priced cost gives you a concrete argument to propose something shorter or more targeted.
What you get back
- The decision or output that would justify the meeting.
- Its async version, drafted, if one is possible.
- The minimum attendee list and shortest workable length.
- A priced fully-loaded cost, and a one-line verdict.
Real example: recurring meeting → verdict
Input: Maya, who runs Brightside, is assessing an inherited slot. The inputs, as-is:
purpose: Monday team meeting, round-table of what everyone's doing
attendees: all 8 at the agency
length: 60 min
recurrence: weekly
Output, the part that matters:
1. Justifying decision: none clear. The round-table informs, it decides nothing.
2. Possible async version: yes. Everyone posts their 3 lines (done / in progress / blocked) in a Slack thread Monday morning. Blockers that affect someone get handled in replies.
3. If kept: bring it down to 20 min, and only invite those with a real call to make.
4. Estimated fully-loaded cost: 8 attendees × 1 hr at $75/h (no hourly cost given, assumption stated) = $600 per week, roughly $2,400 per month.
5. Verdict: make async. An informational round-table at $600 a session does not justify blocking 8 calendars; the Slack thread carries the same information.
Notice what the AI did not do: it invented no hourly cost, it used $75 and flagged it, as the prompt requires. The verdict follows from the number, not from a bias. A meeting carrying a real decision would have come out “keep”.
The full playbook, with the decision grid and examples of meetings kept, shortened, or dropped, ships inside the kit.
Frequently asked questions
- Does this prompt work with Claude or Gemini?
- Yes. It relies on no ChatGPT-specific feature: paste it into Claude, Gemini, or Le Chat and the assessment follows the same structure.
- How does the prompt calculate the cost of the meeting?
- Attendees multiplied by length, at an average hourly cost. If you give none, it uses $75/h and says so plainly, so the number stays transparent and adjustable.
- Does the prompt always push to cancel the meeting?
- No. It returns an honest verdict from four: keep, shorten, make async, or cancel, with one line of reasoning. A useful meeting comes out as one.