Decision Meetings SKILL.md

This is the exact file agents read, including the credit header and licence notice. Save it as decision-meetings/SKILL.md.

---
name: decision-meetings
description: "Run meetings that end with a decision: a framed question, one named decider plus recommender, consulted and informed roles, reversibility, a one to two-page decision memo with options and a recommendation, written input beforehand, a reading-first meeting structure, a decision record and disagree-and-commit follow-through. Use whenever a decision keeps getting postponed, someone needs to run a decision meeting, write a decision memo or one-pager, clarify who decides, or record and communicate a decision."
license: CC-BY-4.0
metadata:
  author: Afeez Adebayo
  author_url: https://afeezadebayo.com
  author_linkedin: https://www.linkedin.com/in/afeez-adebayo/
  author_x: https://x.com/za_icon
  source: https://afeezadebayo.com/skills-md/productivity/meetings-calendar/decision-meetings/
  category: productivity
  subcategory: meetings-calendar
  version: "1.0.0"
  created: "2026-10-10"
  updated: "2026-10-13"
---

# Decision Meetings

Many meetings end with "let's take this offline" or "we need more data". The decision drifts to another meeting, then another, while work waits. Decisions stall for predictable reasons: nobody knows who decides, the options are unclear, the information arrives during the meeting, or people confuse being consulted with having a veto. Clear decision rights and a short written pre-read fix most of this. Amazon has described a practice of starting some meetings by silently reading a written memo before discussion. [Reported, Amazon shareholder letters]

This skill designs and runs meetings that end with a decision: framing the question, naming the decision maker and who is consulted, preparing a short decision memo with options and a recommendation, running the discussion, recording the decision with its reason, and communicating it. It also helps decide when a decision does not need a meeting at all.

## Operating principles
1. **One question, one decider.** Name the decision and the person (or group) who makes it before the meeting.
2. **Read first, then discuss.** A short memo with options and a recommendation, read before or at the start.
3. **Consulted is not a veto.** People give input; the decider decides.
4. **Reversible decisions move fast.** Spend time in proportion to how hard the decision is to undo.
5. **Write it down.** Decision, reason, owner and review date, shared the same day.

## Decision roles
| Role | Meaning |
|---|---|
| Decider | Makes the call and is accountable |
| Recommender | Prepares options and a recommendation |
| Consulted | Gives input before the decision |
| Informed | Told the outcome afterwards |
| Agrees (rare) | Must sign off, such as legal or finance for regulated matters |
Frameworks such as RAPID or RACI use similar roles; the point is that everyone knows theirs.

## When to use, and when not to
Use it for decisions that affect several people or teams: priorities, budgets, launches, hires, vendors, policies, strategy choices and recurring disagreements.

Don't use it for decisions one person can make alone, or for brainstorming (`workshop-facilitation`).

## Inputs
- The question to decide and the deadline.
- The people involved and their roles.
- Options known so far, data and constraints.
- How reversible the decision is.

## Workflow

### Step 1. Frame the question
A single question with a clear answer: "Do we launch on 21 October or 4 November?" not "Launch discussion".

### Step 2. Assign roles
Decider, recommender, consulted and informed. Share them with the invite.

### Step 3. Judge reversibility
Easy to reverse: decide quickly, possibly without a meeting. Hard to reverse: more input and analysis.

### Step 4. Write the decision memo
One to two pages (template below), sent at least 24 hours ahead or read in the first ten minutes.

### Step 5. Gather input before the meeting
Consulted people comment on the memo in writing, so the meeting focuses on real disagreements.

### Step 6. Run the meeting
Read or recap the memo, clarify questions, hear concerns, check the decision criteria, then the decider decides or sets a specific date for the decision with what is missing.

### Step 7. Record and communicate
Decision, reason, options rejected, owner of next steps, review date. Share with informed people the same day.

### Step 8. Commit and review
Once decided, everyone supports the decision even if they disagreed ("disagree and commit"), and it is reviewed on the agreed date.

## Decision memo template
```
Decision needed: <question> by <date>
Decider: <name> | Recommender: <name> | Consulted: <names>
Context: <3 to 5 sentences>
Criteria: <what matters most, e.g. revenue, risk, time, cost>
Options
1. <option> | pros | cons | cost | risk
2. <option> | pros | cons | cost | risk
3. Do nothing | pros | cons
Recommendation: <option> because <reason>
Reversibility: easy / hard
What we need from this meeting: <decision or input>
```

## Meeting structure (30 to 50 minutes)
| Time | Step |
|---|---|
| 0 to 10 | Silent reading or quick recap |
| 10 to 20 | Clarifying questions |
| 20 to 35 | Concerns and trade-offs against the criteria |
| 35 to 45 | Decision, or a dated plan to decide |
| 45 to 50 | Next steps, owners, communication |

## Output format
```
1. Framed question and roles
2. Decision memo
3. Meeting agenda
4. Decision record (after the meeting)
5. Announcement message
```

## Decision record
```
| Date | Decision | Options considered | Reason | Decider | Next steps owner | Review date |
```

## Decisions without a meeting
Many decisions can be made in writing: share the memo, give consulted people a deadline to comment (for example 48 hours), and let the decider post the decision in the same thread. Use a meeting only when there is real disagreement, high stakes or a need to build shared understanding.

## Criteria first
Agree the criteria before discussing options, for example customer impact, revenue, cost, risk and time to deliver, with rough weights. Arguments then focus on how each option scores rather than on preferences, and the decision becomes easier to explain.

## Announcing a decision
```
Decision: we will <decision>, starting <date>.
Why: <one or two reasons>.
What changes: <who does what differently>.
Owner: <name> | Review: <date>
Questions: <where to ask>
```
Send it to everyone informed on the same day, so the decision is not passed on by rumour.

## Using an AI agent
The agent can draft the memo from documents and threads, build the options table, collect written comments into a summary of agreements and disagreements, write the agenda, record the decision and draft the announcement. It does not make the decision.

## Reviewing decisions
On the review date, check the outcome against the reason recorded: did the expected result happen, and would the team decide the same way with what it knows now? Note the lesson in the decision log. Over time, the log shows which kinds of decisions the team makes well and where more input or data usually helps.

## Edge cases and judgement calls
| Situation | What to do |
|---|---|
| Nobody wants to be the decider | Escalate to the person who owns the outcome |
| Missing data | Decide what data would change the decision; set a date |
| Strong disagreement | Record dissent; the decider decides; commit |
| Decision keeps being reopened | Point to the record and review date; reopen only with new information |
| Small, reversible choice | Decide in writing without a meeting |
| Political or sensitive decision | Consult key people one to one before the meeting |

## Handoffs to related skills
- Agendas: `meeting-agenda`
- Notes and decision logs: `meeting-notes-actions`
- Follow-up messages: `meeting-follow-up`
- Prioritisation methods: the Planning and Task Management `prioritisation` skill
- Decision frameworks in depth: the Thinking and Decisions skills

## Worked example
A marketing and product team had discussed a pricing change in four meetings without a decision. The skill framed the question ("Introduce a new mid-tier plan in Q1: yes or no?"), named the head of product as decider, wrote a two-page memo with three options, criteria and a recommendation, and collected written comments from sales and finance beforehand. The 40-minute meeting began with reading, surfaced one real concern about support load, and ended with a yes, a support plan owner and a review date three months later. The record went out the same afternoon.

## Common mistakes
- No named decider.
- Information presented for the first time in the meeting.
- Treating every opinion as a veto.
- Ending with "let's discuss next week" without a date or missing input.
- Not recording why the decision was made.

## Self-check before you answer
- Is the question framed with a clear answer?
- Are roles named, with one decider?
- Is there a memo with options and a recommendation?
- Is the decision recorded with reason, owner and review date?

---
Created by Afeez Adebayo, Marketing, SEO & Growth Expert
Website: https://afeezadebayo.com
LinkedIn: https://www.linkedin.com/in/afeez-adebayo/
X: https://x.com/za_icon
Source: https://afeezadebayo.com/skills-md/productivity/meetings-calendar/decision-meetings/

Licence: CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). You may copy, share and adapt this skill, including for commercial use, as long as you credit Afeez Adebayo and link to the source URL above. Keep this notice in any copy or adapted version.