Write a brief that survives handover

Three people, and only one of them is you.

The person who inherits the account in four months, who needs to know why the split is what it is before they change it. The person approving a change against it, who is deciding whether what they are looking at matches what was intended. And an agent, which reads the brief when you ask it to help with the items and has no other source for your intent.

Everything below follows from that list. A brief written for the person who wrote it is a set of reminders. A brief written for the three above is a set of decisions with their reasons attached.

What this is for. One sentence, in business terms rather than advertising terms. "Fill the January intake" rather than "drive conversions". Somebody reading it in April needs to know whether the thing still exists.

Who it is aimed at. Who they are, not which targeting settings you used. Settings are on the campaign and change; the description of the person does not, and it is what lets somebody judge whether a new setting is still aimed at the same audience.

What counts as working. Stated as a comparison where possible — better than what it replaced, over a window long enough to mean something. An absolute number with no basis becomes a target people defend, and a defended number outlives its usefulness by months.

What has already been decided against. The field everybody skips and the one that saves the most time. Two options considered and rejected, with one clause each on why. Without it, the person inheriting the campaign re-runs the same argument in week two and usually reaches the opposite conclusion for reasons nobody recorded.

The constraints that are not negotiable. Compliance limits, a claim the business cannot make, a landing page that will not change before a date, a creative that has to be used. These are the things that look like poor judgement to a newcomer and are not.

What happens at the end. Whether the flight stops, extends, or feeds something else. A campaign with no stated ending is a campaign that gets paused by whoever notices it first, three weeks after it should have been.

The review date. One date, written before the flight starts. The plan does not watch anything for you — it states the intent and the reporting side states the outcome, and holding the two against each other happens on the date you wrote down.

Anything time-relative. Next month, recently, the current quarter. A brief read in April cannot resolve any of it. Write dates.

Anything about the tooling. Which screen you used, which agent you asked, which report you exported. None of it is about the campaign and all of it will be wrong within a year.

Numbers with no denominator. A target return with no margin behind it, a cost per result with no window, a budget with no window. Each becomes a number people quote and nobody can check. If you want the arithmetic in the brief, put the inputs in it too — the ROAS calculator and the CPA calculator work the same figures from your own margin, so the inputs are what is worth recording.

Tone. Confidence about a hypothesis reads as fact to whoever arrives later. Say which parts are believed and which are being tested; the brief is the only place that distinction survives.

In the plan, in the brief field. A media plan carries a name, a brief, a window, one currency and one line per campaign you intend to run, and nothing in it touches live spend — which makes it the right place for reasoning that has to survive, because the document is not a campaign and does not get paused.

While the plan is a draft everything in it is free to edit, so the brief can be sharpened as the plan is argued about. Once the plan has been materialised it is frozen — name, brief, window, currency and items all stop being editable — because the plan has become the record of what was drafted, and a record you can edit behind is not a record. That freeze is the reason to write the brief properly before materialising rather than after.

The campaign brief template is the long form for the parts of an organisation that need a document rather than a field, and the media plan template carries the same structure as a sheet.

The record of what was actually done. That exists already and it is better than your memory: every applied change is a row naming who acted, which authority cleared it, and what it moved from. A brief that tries to be a change log is a brief that goes stale.

Which is the right moment to state plainly what the product does to an account, because a brief is where somebody often assumes more. Your connections read your accounts. A change anyone asks for is written as a draft on the record, with the state it would change from and the checks it passed. Applying it is gated on approval — by you, or by a rule you wrote — and no platform write reaches a customer account until that platform has granted write access. Applying a cleared change to a platform is in private beta.

So the brief carries intent, and the record carries events. Keeping those separate is what stops either from being rewritten to agree with the other.

Hand it to somebody who has never seen the account and ask them two questions. Why is the budget split the way it is. What would have to be true for you to change it.

If they can answer both from the document, it survives handover. If they cannot, the missing paragraph is almost always the one about what was decided against.

How to write a media plan is the document this brief sits inside, and Plan a product launch flight is the case where the brief has to carry three phases with three different definitions of working.