A good brief is short and hard to write. It states a problem, the evidence for it, who has it, what a win looks like, and what is genuinely fixed. Everything else is negotiable, and saying so is what lets a studio bring you something better than you asked for.
Lead with the problem, not the solution
"We need a new website" is a solution. "Enterprise buyers cannot tell within thirty seconds whether we serve their sector, and sales spends the first call explaining it" is a problem. The second version can be solved several ways, at least one of which is cheaper than a new website.
If the brief names the deliverable but not the problem, you have pre-committed to the answer and hired us to render it.
What to include
- The problem, in one paragraph, with the evidence you already hold.
- Who it affects, specifically. Not "users": the two or three real segments and which one matters most.
- What success looks like in numbers, even rough ones. "Fewer support tickets about billing" is measurable enough to start.
- The constraints that are real: launch dates tied to events, technology you cannot leave, regulation, languages, brand rules that are truly fixed.
- What is out of scope, and what you have already tried and rejected. This saves everyone a month.
What to leave out
A list of sites you like, unless you can say what specifically about each one applies to your problem. Detailed feature specifications, unless they are contractual. And any adjective that could describe every company in your sector: modern, clean, premium, user-friendly.
A test before you send it
Give the brief to someone in your company who does not work on the project. If they cannot tell you who the customer is and what would count as a win, it is not ready. If they can, you will get proposals that differ in approach rather than in price, which is the point.