How to write a brief

Eighty percent of people believe they write good briefs. Ten percent of agencies agree.

Those figures come from the largest global study on the subject, across more than 1,700 respondents in 70 countries. The gap widens on the detail. Seventy-eight percent feel their briefs give clear strategic direction, against just 5 percent of agencies. Eighty-three percent believe their language is clear: only 7 percent of agencies agreed.

The interesting part is who is writing them. Almost three-quarters have more than ten years of experience and two in three hold a senior or director title. So this isn’t a competence problem, it’s an information problem with one cause: a brief is written from inside a business by someone who knows everything, then read from outside by someone who knows nothing.

Here’s how we can close that gap.

Before you write anything

The natural instinct is to write for the agency or supplier you already work with, the incumbent. They know the politics. They remember what happened last year. They can fill in the blanks without being asked.

Here’s the rub: every blank you leave for them is a hole for everybody else.

So try this before you start. Picture the agency you would most love to work with reading this cold, with no relationship, no history, no back-channel. What is everything they need to know to give you their best possible thinking? Answer that question honestly and most of the brief writes itself. Skip it and you have written a document only the incumbent can answer well, which rather defeats the purpose of running a process at all.

What exactly are you asking for?

An RFI asks who is out there. An RFP asks for an approach and a price. An RFQ asks for a price against a scope you have already defined. These are three different briefs asking three different questions.

Label the brief accurately and you get answers to the question you actually asked. Call an RFP an RFI and you will receive credentials decks when you wanted thinking, then lose two weeks going back asking for an actual concept.

Then there’s the balance of complexity versus brevity. Creative professionals need detail, what we don’t need is an archaeologist to excavate it from 100 pages of A4. The brief that buries its point almost always produces proposals that miss it. David Ogilvy's observation applies: big ideas are usually simple ideas. If your brief cannot be understood in one read by someone outside your industry, it needs another pass.

Five tests

The SMART framework – Specific, Measurable, Achievable, Relevant and Time-Bound – is a way of writing objectives that people can actually act on. It transfers to briefs almost perfectly.

SPECIFIC

Open with who you are. History, mission, market position, what genuinely makes the organisation different. Assume no prior knowledge. The agency can research you, though what they cannot research is how you see yourselves.

Then state the problem, the fundamental issue sitting beneath everything. "We need a brand experience" is an activity. "Our customers do not renew after year one" is a problem. It opens the door to answers you had not considered.

Be exact about scope. Which tasks, which deliverables, which technical requirements, which milestones. Say whether you want strategy only, strategy through to production, or full onsite delivery. State clearly how much creative latitude exists: are you applying existing brand collateral or building something new?

Show your work. Use reference images: previous experiences, an event you admired, a creative concept you felt got it right. What you show is almost always more useful than what you say.

Vague language often does the most damage. Stephen King wrote that the road to hell is paved with adverbs. Alas, briefs can often be full of them. Truly innovative. Highly engaging. Completely immersive. When you write innovation, define it, because it means something different to everyone reading it. Some will hear technology. Some will hear format. Some will hear risk.

MEASURABLE

Define what success looks like in numbers you will genuinely check afterwards. Attendance, conversion, ROAS, retention, share of voice, NPS, pipeline, PR. Who owns the KPIs, who is defining them? If a key goal cannot be measured, say so openly and explain how you will judge it instead.

For the brief itself, be crystal clear about evaluation criteria. If the split is 80 percent technical and 20 percent commercial, say so. If cultural fit carries weight, say that too. Agencies build proposals to the scoring model, so an unpublished model produces work aimed at a target that is unseen. It becomes a guessing game. Publishing it costs you nothing and improves every submission you receive.

ACHIEVABLE

Please, please, please, please, please state the budget, or, at a minimum, outline a genuine range. This is the most contested line in any brief and the one that saves the most time (and, ultimately, money), for both the client and the agency. Creativity sits hand in hand with cash, which is the real craft in this work: pushing boundaries as far as possible against an agreed ceiling. Otherwise, come the submission deadline, you’re sat with multiple proposals answering multiple questions, which is not a comparison: it’s an exercise in futility. Be upfront about the numbers and you will have experience design that doesn’t just meet expectations, it exceeds them.

Submission mechanics belong here too: format, file type, page or size limit, portal or email, plus the person whose inbox it lands in.

RELEVANT

Context is king. Is this a new experience, or something that has been in place for years? Share what has happened before. What worked, what didn’t, what the KPIs demonstrated, what you would never repeat. Previous delivery data is the single most useful attachment that the vast majority of briefs never include.

Is there an incumbent? Say so, we’re going to find out anyway. Hearing it from you signals that this is a process worth entering, that you’re transparent and open right from the start of the relationship.

Be honest about appetite. "We were happy with last year, but please evolve it" and "we want something new" look nearly identical on paper but can produce completely different work.

TIME-BOUND

What are the next steps? Publish the full sequence. The brief meeting, the deadline for clarification questions, the deadline for submission, shortlist notification, presentation date, decision date, project start. A brief that gives only a submission date tells an agency nothing about how to resource the work.

Before the brief hits the portal or inbox, interrogate whether those dates are realistic. Rushed deadlines produce rushed proposals. You have to live with what comes back and two extra weeks at the front costs you far less than a compromised twelve months at the back.

State clearly what happens after submission, including what feedback you will provide, win or lose. Feedback is the cheapest thing you can offer and the thing agencies value most. It costs an hour and will be remembered long after the decision.

Why all of this matters

For both clients and agencies, a significant amount of time, resource and money is lost to poor briefs and misdirected work. That waste never appears as a line item: it surfaces as extra rounds, revised submissions, work that misses, relationships that can sour.

Across the entire project process, the brief is the key deliverable that every subsequent decision refers back to. Every proposal, every concept, every production choice, every discussion about scope, traces back to what was written at the start. An extra hour or two spent on clarity now is worth a hundred more spent on course correction afterwards.

The bottom line? Nobody has ever complained about a brief that was too clear.