What a memo does that a deck cannot
Slides let a presenter control pace and skate past weak points; a memo sits still and takes questions. Writing the decision down in prose exposes gaps that survive perfectly well in bullet points, which is exactly why the discipline is worth the discomfort. If the memo cannot be written, the decision is not ready to be made.
The memo skeleton
| Memo section |
How to write it |
| Context |
Three sentences maximum on why this decision exists now. If the reader needs more background than that, attach it. Do not open with a history lesson. |
| Decision requested |
One sentence, specific enough to be minuted: what is being decided, for how much, by whom, effective when. If you cannot write this sentence, stop drafting and go find out. |
| Options considered |
Two to four real options including "do nothing" or "do less". For each: one line on what it is, one line on why it was or was not chosen. Options added to make the preferred one look good will be spotted. |
| Evidence |
The three to five facts the recommendation rests on, each with its source and age. Distinguish what is known from what is believed: approvers forgive uncertainty, not disguise. |
| Risks |
The two or three ways this goes wrong that would actually matter, each with the mitigation and its cost. A risk list with no expensive mitigations means the risks were not taken seriously. |
| Dependencies |
What must be true or delivered elsewhere for this to work: other teams, systems, approvals, market conditions. Name the owner of each dependency. Unowned dependencies are how approved decisions stall. |
| Recommendation |
The author's actual position, stated plainly, with the one consideration that most nearly changed it. Hedged recommendations transfer the thinking back to the room and waste the memo. |
| Reversibility |
What it costs, in money and time, to undo this after approval, and the last date at which changing course is still cheap. This single section changes how much scrutiny the decision deserves. |
| Decision and conditions (completed in the meeting) |
The outcome as minuted: approved, rejected or deferred, any conditions attached, and the named owner of each condition. Left blank on circulation; filled in before the meeting ends. |
Circulation and timing
Send the memo at least three working days before the meeting and open the meeting by asking who has read it. Silence answers the question. The author is whoever wants the decision made; the memo is theirs alone to draft, though the evidence section should survive a hostile check by someone who did not write it. Decisions of consequence deserve ten minutes of silent reading at the top of the meeting rather than a re-presentation of what was circulated.
How good memos go wrong
- The decision requested is written as a theme ("align on our vendor approach") rather than a commitment, so the meeting discusses instead of deciding.
- Reversibility is omitted because the author fears it invites second-guessing, when in practice it is what earns fast approval for reversible choices.
- The memo grows past two pages as stakeholders each add a defensive paragraph, and the added length dilutes the ask it was built around.
- The final section is never completed, so six months later nobody can establish what was actually approved or on what conditions.
When the memo deserves a hostile reader first
For decisions that are expensive to reverse, the cheapest insurance is a reader with no stake in the outcome seeing the memo before the approvers do. Senior operators who have made the same class of decision will find the soft evidence and the unpriced risk in an afternoon. It is considerably better to hear it from them, confidentially, than from your board. That pre-read is what a client brief to Digital Advisory's Global Board provides.