How to write a marketing brief you can measure

Marketing work exists to change something: the number of qualified conversations, the mix of enquiries, the price buyers expect, or how quickly a segment recognises the offer. Naming that change first determines what the brief should contain. A brief written to support "more awareness" will produce work that cannot be evaluated, because awareness has no agreed evidence in most businesses.

A useful test: if this work succeeded, what would the business do differently next quarter? If the answer is nothing, the brief is describing activity rather than a purpose.

1. The problem, in the buyer's terms. What is happening now that makes the work necessary, and what that costs the buyer.
2. One audience. A role with a problem, not a market segment. Where two audiences need different claims, that is two briefs.
3. The claim. One sentence the buyer could repeat. If marketing, sales, and the offer page would phrase it differently, it is not yet a claim.
4. The response. What the buyer is asked to do, and what happens when they do. Response definitions matter more than reach figures because they determine what counts as an enquiry.
5. The measure and baseline. The figure that should move, its current value, and the window in which it will be reviewed.
6. Constraints. Budget, timing, approvals, brand rules, and any claim that legal or compliance has already ruled out.
7. Assumptions. What the plan depends on being true, and what would show it is not.

• Response over reach. Impressions describe delivery; responses describe effect. Reach can be reported alongside, but not as the measure.
• Quality defined in writing. A qualified conversation needs a definition that sales and marketing both accept, or the two will disagree about the result.
• Baseline before launch. A figure recorded after the work has started is usually chosen to flatter it.
• One primary measure. Secondary measures explain the primary one; a brief with five primary measures cannot be judged.

Where the numbers depend on pipeline stages, the stage definitions have to exist first. The notes on a repeatable sales process cover what a stage needs before it can be counted.

Write three outcomes before work starts: the result that means continue, the result that means adjust, and the result that means stop. This is the part most briefs omit, and it is the part that prevents a review becoming a negotiation about interpretation.

It also makes the brief useful to the people doing the work. A team told what would count as failure can make better decisions inside the work, rather than waiting to be told afterwards.

The example below is hypothetical and describes a pattern rather than a client engagement.

• Attribution is imperfect. A buyer who read one article and replied to an unrelated email will be counted inconsistently. The brief should state how response will be attributed rather than assuming precision.
• Short windows favour direct response. Work that builds recognition over months will look weak against a four-week measure. Where the objective is genuinely long-term, the review point has to match it.
• Briefs constrain creativity. Some of that constraint is the point; some can crowd out an idea that would have worked. Keeping the brief to the problem and the claim leaves the execution open.
• A good brief does not fix a weak offer. If the claim cannot be evidenced, the brief will document that rather than resolve it.

Briefing and measurement sit inside the Marketing, Media & Digital capability, alongside positioning and channel decisions. Where the question is whether to build internal capacity or use an agency, the comparison of a fractional CMO and an agency covers the structural choice.