The Recurring Meeting Auditor: Decide Which Standing Meetings Still Earn Their Slot on the Calendar
Why this prompt matters
A mid-size team carrying even 3 unnecessary weekly meetings at 5-8 attendees and 30-45 minutes each burns 8-12 hours of collective team time every single week — 400-600+ hours a year, the rough equivalent of a quarter of a full-time employee's annual hours, spent on meetings that evaporate the moment someone asks "what decision does this produce?" and nobody has a clean answer.
What we use it for
Your 12-person engineering team has 6 standing meetings on the calendar, several inherited from a reorg 18 months ago. Nobody fully remembers why half of them exist, and two engineers have separately mentioned losing a full morning a week to meetings. You need an objective, defensible way to decide what to cut without personally arguing the case in a room full of people who each created one of these meetings.
Prompt
Role: Act as an operations efficiency consultant who has audited meeting calendars at dozens of companies and specializes in identifying "zombie meetings" — recurring meetings that have outlived their original purpose. Context: Team/company context: [TEAM SIZE AND FUNCTION, e.g., "12-person engineering team"]. Recurring meetings list — for each one, provide: name, frequency, duration, number of attendees, who created it and roughly when (if known), and its stated purpose (if known, otherwise write "unknown"): [LIST YOUR RECURRING MEETINGS, ONE PER LINE, WITH AS MUCH OF THE ABOVE AS YOU KNOW] Task: For each meeting, score it against four criteria: 1. Decision-making value — does this meeting produce a decision that couldn't be made async? 2. Information transfer that could be async — is most of the content a status update that could be a written doc or message instead? 3. Attendee necessity — are all attendees actually needed for the full duration, or could most of it be a smaller group? 4. Staleness — does the stated (or inferred) original purpose still match what the team actually needs today? Based on the scoring, output a verdict for each meeting: KEEP AS-IS / SHORTEN / MAKE ASYNC / KILL, with one or two sentences of reasoning, and — for SHORTEN or MAKE ASYNC verdicts — a specific suggested replacement (e.g., "a shared doc updated by Friday EOD" or "a 15-minute version with only the 3 people who make the decision"). Constraints: Do not recommend killing meetings that involve legally required reviews, client-facing calls, or 1:1s unless the input explicitly flags them as low-value. Be specific about what exactly becomes async — never just write "make this async" without naming the replacement format. Identify the single highest-value meeting on the list and explicitly flag it as one to protect from any future cuts. Output Format: ## Meeting-by-Meeting Verdicts [Table or list: Meeting Name | Verdict | Reasoning | Replacement (if applicable)] ## Protect This One [Name the single highest-value meeting and why] ## Time Reclaimed [If every SHORTEN and KILL recommendation is implemented, calculate total attendee-hours per week reclaimed, using the duration and attendee count provided for each affected meeting]
Result
Meeting-by-Meeting Verdicts
| Meeting | Verdict | Reasoning | Replacement |
|---|---|---|---|
| Monday Status Sync (30 min, 8 people, weekly) | MAKE ASYNC | Each person reports what they did last week with no discussion or decisions made — pure status broadcast that doesn't need real-time attendance from 8 people. | A shared doc or Slack thread where each person posts 2-3 bullet points by Monday 10am. |
| Design Review (45 min, 6 people, weekly) | SHORTEN | Real decisions get made here, but only 3 of the 6 attendees speak in a typical session — the other 3 are there "for visibility." | 25-minute session with the 3 decision-makers; others get the decision summary afterward. |
| All-Hands Engineering (60 min, 12 people, weekly) | KEEP AS-IS | Genuinely serves cross-team alignment and surfaces blockers that would otherwise stay siloed — attendee feedback in the input confirms this is considered useful. | — |
| Retro (30 min, 8 people, bi-weekly) | KEEP AS-IS | Produces concrete process changes most cycles; this is decision-making work, not status reporting. | — |
| Legacy "Project Phoenix" Check-in (30 min, 5 people, weekly) | KILL | Project Phoenix shipped four months ago. Nobody on the list could state its current purpose when asked; this is a textbook zombie meeting. | None needed — cancel outright. |
Protect This One
The bi-weekly Retro is the highest-value meeting on this list — it's the only recurring meeting that reliably converts discussion into a tracked action item, and teams that cut retros under time pressure tend to lose the only structured mechanism they have for catching process problems before they compound.
Time Reclaimed
Making Monday Status Sync async: 8 people × 30 min = 4 hours/week reclaimed.
Shortening Design Review from 45 to 25 min: 6 people × 20 min = 2 hours/week reclaimed.
Killing the Project Phoenix check-in: 5 people × 30 min = 2.5 hours/week reclaimed.
Total: 8.5 attendee-hours reclaimed per week — roughly 442 hours per year, equivalent to about a quarter of one full-time employee's annual working hours.
Recurring meetings are the subscription services of the corporate calendar: easy to add, almost never canceled, and most people have no idea which ones they're still paying for until someone makes them look at the bill. A meeting created to solve a problem that no longer exists keeps running because canceling it requires someone to publicly declare it's not worth the room's time — and nobody wants to be the one who says that about a meeting their manager, or their manager's manager, started.
The prompt below removes that social cost by making the audit a scoring exercise rather than a personal judgment call. Instead of asking "should we cut the Monday sync," it asks four narrower, harder-to-argue-with questions about each meeting: does it produce a decision, could the content be async, are all the attendees actually needed, and does the original purpose still match current reality. A meeting can fail on all four counts without anyone in the room being wrong to have created it eighteen months ago — circumstances just changed.
Why "make it async" alone isn't a real recommendation
The most common failure mode in meeting-reduction advice is stopping at "this could be an email." That's true of almost every status meeting, and it's also useless as a recommendation, because nobody acts on vague advice. The prompt's constraint section forces a specific replacement format for every SHORTEN or MAKE ASYNC verdict — a shared doc with a deadline, a smaller group with a time limit, a Slack thread with a cutoff — because a named replacement gets adopted and an unnamed one quietly doesn't happen, and the meeting creeps back onto the calendar within a month.
Why the output protects one meeting explicitly
Meeting-cutting exercises have a failure mode in the opposite direction too: once a team starts applying scrutiny, it's tempting to keep cutting past the point of diminishing returns, including meetings that are quietly doing real work. The "Protect This One" section forces the audit to name the single highest-value meeting on the list and explain why, so the exercise doesn't become indiscriminate calendar-clearing that removes the one structured mechanism a team has for catching problems — a retro, a design review, a planning session — along with the genuinely dead weight around it.
How to use this well
Be honest in the input about attendee count and duration — the final "time reclaimed" number is only as credible as the inputs, and a number built on optimistic estimates won't survive someone in the room doing quick mental math and finding it wrong. If you genuinely don't know a meeting's original purpose, write "unknown" rather than guessing a charitable one; an honest "unknown" is itself useful signal that a meeting has drifted from its reason for existing, and the audit should treat it that way rather than inventing a justification the meeting's current content doesn't actually support.