A brief in plain language: not "describe your target audience" but what the site must make happen, who decides, and what already exists. Plus the three answers that move the price more than anything else.
Most briefs we receive are a list of pages. Most projects that go well started with something else: a short, unglamorous set of answers that took an afternoon to write and saved a month.
You do not need a document. You need to have decided these things, and to have written them somewhere a supplier can read.
One sentence, with a number in it if possible. Not “represent the company professionally” — that is a feeling, and feelings cannot be delivered or measured. Something closer to: twenty qualified enquiries a month; orders that no longer need a phone call; a self-service area that removes the Monday morning email pile.
This sentence does more work than the rest of the brief combined. It decides what gets built, what gets cut when the budget tightens, and how anyone will know afterwards whether the money worked.
Name one person who can approve without convening anybody. Not who has opinions — who signs off.
If the honest answer is “four of us together”, that is fine, but it is a fact about the schedule and the price rather than a detail. Group approval is slower and more expensive, and the difference is measured in weeks. Say it at the start and a good studio will design the process around it, with fewer and better-prepared decision points.
Four questions, answered honestly.
There is a real difference between “we would like it by autumn” and “our trade show is on 14 October and the URL is already printed”. The first is a preference; the second is a constraint that reshapes the whole plan.
Say which one it is. Studios treat invented deadlines differently once they discover they were invented, and treat real ones properly when they know early — including by telling you which parts to drop to make the date.
Withholding the budget feels like negotiating leverage. It mostly wastes everybody's time: you receive proposals aimed at different projects and cannot compare them, and the studio designs a scope that may be twice or half what you can afford.
A range is enough. It lets a serious supplier tell you what is achievable inside it, or say honestly that it is not — which is more useful than a proposal engineered to fit a number you never named. We wrote separately about where that money actually goes.
Worth planning for, because it lands on your calendar rather than the studio's: a couple of conversations of an hour or two, access to whoever actually handles orders or enquiries today, a look at the numbers you already have, and a straight answer about what has failed before.
That last one saves the most. Every business has a previous attempt — an agency that went quiet, a rebuild that never launched, a tool nobody adopted. Telling that story early stops it repeating and is not embarrassing; it is the most useful hour in the project.
Out of everything above, three dominate: whether your content exists, how many people approve, and how many systems you do not control. Two projects with identical page counts can differ by a factor of three on those three answers alone.
If you want a cheaper project, work on those before you work on the brief.
What happens if we do nothing? Sometimes the honest answer is “not much for another year”, and that is worth knowing before spending. Sometimes it is “we keep losing enquiries to a competitor whose site loads”, which changes both the urgency and the budget.
Who inside the business will own this after launch? A site with no owner drifts: nobody publishes, nobody watches the numbers, nobody notices the form broke. If the answer is nobody, either appoint someone or buy an arrangement that covers it — otherwise you are funding an asset that starts decaying on day one.
Resist specifying the technology, the content system or the design details. Not because your opinion is unwelcome, but because those are answers to questions you are paying somebody else to ask properly, and fixing them early removes the cheaper options before anyone has looked.
The exception is a genuine constraint: your team already runs one platform, an internal policy requires EU hosting, a compliance rule limits where data lives. Those are requirements, and they belong in the brief. “We heard good things about that framework” is not.
If writing a brief feels like a project of its own, send this instead — it is enough to start a real conversation:
What we do and who buys it. What this site has to make happen, in one sentence with a number. Who approves work. What content exists today and what does not. Which systems it must connect to. When it needs to be live and why that date. Roughly what we can spend. What we already know is wrong with what we have.
Eight answers. An afternoon. It will save more time than any other document in the project, and it doubles as the thing you compare proposals against later.
Answers like these should produce more than a price. A useful reply pushes back somewhere: questions the outcome sentence, suggests something smaller, names a risk you had not considered, or says plainly that part of the plan is not worth doing.
A proposal that agrees with everything you wrote is a quote, not a partner. And when the quotes do arrive, the eight lines that decide everything matter more than the totals at the bottom.