AIO Apex
en
Claude Sonnet 4.6 -- handles the tone shift across all three escalation tiers and the conditional 'flag, don't invent' logic for undated items reliably. GPT-5.4 and Gemini 3 Pro also work well; smaller or budget models tend to collapse all three tiers into the same tone, or quietly assign a deadline to a flagged item instead of surfacing it as an open question.A product team just wrapped a 45-minute sprint planning meeting with eight action items split across three owners and zero written follow-up system. The project lead has two other launches to run this week and can't personally remember to chase each person at the right moment.Writing & Communication

The Follow-Up Enforcer: Turn a Meeting's Action Items Into Deadlines Nobody Can Quietly Ignore

Share:
The Follow-Up Enforcer: Turn a Meeting's Action Items Into Deadlines Nobody Can Quietly Ignore

Why this prompt matters

Meetings are good at producing commitments and bad at tracking them, because the follow-up step itself usually has no owner. When a slipped deadline goes unflagged, the typical outcome isn't a quiet delay -- it's a second meeting to re-decide what the first meeting already settled, plus whatever schedule risk compounds downstream while nobody notices the item stalled.

What we use it for

A product team just wrapped a 45-minute sprint planning meeting with eight action items split across three owners and zero written follow-up system. The project lead has two other launches to run this week and can't personally remember to chase each person at the right moment.

Prompt

Act as an operations-minded [EXECUTIVE ASSISTANT / PROJECT MANAGER / TEAM LEAD] who specializes in converting verbal meeting commitments into a system nobody can quietly ignore.

CONTEXT:
Meeting: [MEETING NAME/PURPOSE]
Date: [MEETING DATE]
Attendees and their roles: [LIST NAMES AND ROLES]
Typical slip tolerance before a delay becomes a real problem on this team: [NUMBER] business days
Raw notes or action items from the meeting: [PASTE MEETING NOTES OR BULLET LIST OF COMMITMENTS HERE]

TASK:
1. Extract every discrete action item from the notes. For each one, identify: the owner, the specific deliverable (not just the topic), and the deadline -- either stated explicitly or your best inference from context (e.g. "before the Friday launch" -> a specific date).
2. For any action item with no stated or inferable deadline, flag it as "NO DEADLINE SET" instead of guessing one.
3. For each action item WITH a deadline, draft three follow-up messages at escalating urgency:
   - Tier 1 (friendly check-in, sent 1 day BEFORE the deadline)
   - Tier 2 (direct nudge, sent on the day the deadline passes with no update)
   - Tier 3 (manager-escalation note, sent [NUMBER] business days after Tier 2 if still no update)

CONSTRAINTS:
- Tier 1 must read as a casual, low-pressure Slack-style message -- assume good faith.
- Tier 2 must be direct but not accusatory -- state the deadline, ask for a status, no apology-padding.
- Tier 3 must be factual and unemotional, written as an email, and must cc [MANAGER NAME/ROLE] -- never cc a manager in Tier 1 or Tier 2.
- Keep every message under 80 words.
- Do not invent commitments that are not in the source notes.
- Do not assign a deadline to an item flagged "NO DEADLINE SET" -- surface it as an open question for the [MEETING OWNER] to resolve instead.

OUTPUT FORMAT:
Part 1 -- Action Item Table: columns Owner | Deliverable | Deadline | Status (flag "NO DEADLINE SET" where relevant).
Part 2 -- For each action item WITH a deadline, three labeled message drafts (Tier 1 / Tier 2 / Tier 3) in the channel specified above.

Result

Part 1 -- Action Item Table

OwnerDeliverableDeadlineStatus
PriyaFinalize API documentation for the v3 endpointsFriday, Nov 20On track
MarcusRewrite pricing page copy for the new tier structure—NO DEADLINE SET — flagging for Jordan to resolve before launch

Part 2 -- Follow-Up Drafts for Priya's Item

Tier 1 (Slack, sent Thursday Nov 19):
"Hey Priya! Quick check — how's the API docs update for the v3 endpoints coming along? Just making sure nothing's blocking you before Friday. No action needed if you're on track, just flag me if you need anything."

Tier 2 (Slack, sent Friday Nov 20, deadline day, no update received):
"Priya — following up on the v3 API docs. This was due today and I haven't seen an update. Can you give me a status by end of day? If it's slipping, let me know what's blocking it so we can adjust."

Tier 3 (Email, sent Tuesday Nov 24 — two business days after Tier 2 with still no response, cc Jordan):
Subject: Status needed — API docs for v3 endpoints (overdue since Nov 20)
"Hi Priya, looping in Jordan here. The v3 API documentation was due Nov 20 and I haven't received a status update despite a follow-up on the deadline day itself. Could you share where this stands and a revised completion date? Jordan — flagging in case this affects the launch timeline."

Flagged item: Marcus's pricing page copy has no agreed deadline in the notes. Recommend Jordan confirm one before the launch window closes, since it is currently untracked and will not appear in any of the above escalation tiers until a date is set.

Most meetings produce commitments, not accountability. Someone says "I'll have that by Friday," it goes into nobody's calendar, and by the following week's meeting the same item gets re-discussed from scratch. This prompt closes the gap between "decision made" and "decision tracked" by turning raw meeting notes into a structured follow-up system with three escalating message tiers.

Why a Three-Tier Structure

A single generic reminder treats a one-day slip the same as a two-week ghosting, which either nags people too early or lets real problems fester unflagged. The prompt forces the model to draft three distinct messages per action item: a low-pressure Slack check-in sent before the deadline, a direct but calm nudge sent the day it passes, and a factual, manager-cc'd email reserved for items that are genuinely stuck. Separating these tiers by tone and channel means the first reminder never reads as an accusation, and the escalation to a manager only happens once it's actually earned -- which keeps the system usable without burning social capital every time someone runs a day late.

Why It Flags Missing Deadlines Instead of Inventing Them

Meeting notes are often sloppy -- "Marcus will handle the pricing page" with no date attached. A weaker prompt would guess a deadline and create a false sense that the item is tracked. This one explicitly instructs the model to mark undated items as "NO DEADLINE SET" and route them back to the meeting owner as an open question, rather than fabricating a date nobody agreed to. That distinction matters: a tracked item with a wrong deadline is worse than an untracked one, because it creates confidence where none is warranted.

How to Use It

Paste the raw notes or bullet list from the meeting, name the attendees and their roles, and set a "slip tolerance" -- how many business days a delay can run before it becomes a real problem for that team. The model handles the extraction and drafting; you still decide when to actually send each tier, since the point is to have the right message ready the moment it's needed, not to auto-blast reminders on a timer.

Run this right after any meeting that generates more than two or three action items across different owners. Standups with single-owner tasks don't need the overhead, but cross-functional syncs, launch-readiness reviews, and planning sessions usually do.

workplace communicationmeetingsproject-management
Share: