Write the agency brief before you book the calls

The brief comes before the search

Agency searches often start one step too late. A team asks for names, opens a few directories, and books calls before the project has been described clearly enough for anyone to judge fit. Five agencies then hear five slightly different versions of the same need, and the comparison rests on impressions from the calls rather than on the work.

There is decent evidence that this is expensive. In a global study by the BetterBriefs Project and the IPA, respondents estimated that a third of marketing budgets are wasted through poor briefs and the misdirected work that follows. In the same study, 80% of marketers believed they write good briefs; 10% of agencies agreed.

A useful brief does not need to be a procurement document. It should explain the problem, the constraints, and the decision that has to be made next. A rough one-page version written for this project is a better starting point than a long requirements list copied from an old one.

What to write down first

Start with the business situation rather than the deliverable. A new website, campaign, CRM setup, or product sprint can mean very different things depending on why it matters now. Write the internal trigger plainly: revenue has stalled, support tickets are rising, a market launch is blocked, or leadership needs proof before funding a larger build.

Then add the non-negotiables. Budget range, timing, languages, geography, accessibility requirements, legal or security constraints, and the people who must approve the work all change which providers are realistic. Sharing them early can feel uncomfortable, but withholding them mostly delays the moment a provider tells you the project is not feasible as described.

Finally, separate outcomes from preferences. An outcome is something the work must make possible: fewer manual sales steps, faster pages, better qualified leads. A preference is your current guess about how to get there. Agencies can challenge a preference; they cannot challenge an outcome they have never seen written down.

The GOV.UK Service Manual's guidance on discovery is a useful model here, because it treats early work as a way to decide whether a direction is worth pursuing at all.

How this improves matching

Once the brief exists, the first list of candidates becomes less about who looks impressive in general and more about who fits this particular project. A small specialist may beat a larger agency if the project needs deep sector knowledge. A local partner may matter if workshops and language are central. A delivery-heavy team may be wrong if the open question is still strategy.

Goodfit is designed around that kind of comparison. Provider profiles are more useful when they can be read against a real brief: services, industries, budget bands, location, languages, and proof of similar work.

None of this requires the brief to be perfect. It has done its job when the first conversation is specific enough that both sides can decline quickly, or continue for reasons that hold up after the sales call.