Most briefs are too long
We get briefs that run twenty pages. Company history, competitors, personas, technical notes.
We read all of it. But the useful part is usually four or five sentences buried on page eleven.
A brief is not a place to prove you did your homework. It is a place to say what you want and why.
The six things a brief needs
- What should exist at the end. Be concrete. A logo, a five-page site, a design system.
- Who it is for. One sentence about a person, not a persona document.
- What problem it solves. What is going wrong right now.
- What cannot change. A name, a colour, a platform, a date.
- How you will judge it. What good looks like to you.
- Budget and date. Both real, not aspirational.
That is the whole thing. If you can answer those six, you have a brief.
Brief the problem, not the solution
The most common mistake is asking for a solution instead of describing a problem.
“We need a new homepage” is a solution. “People land on our homepage and still cannot tell what we sell” is a problem.
The second one is more useful, because it might not need a new homepage. It might need a new sentence. Nobody can suggest that if the brief only asks for a rebuild.
Say what you will not change
This section saves the most time and almost nobody includes it.
Tell us what is settled. The name is not up for discussion. The green has to stay because it is on ten thousand printed boxes. It has to launch before the trade show in March.
Constraints are not a problem. Unwritten constraints are, because everyone finds them the expensive way.
How you will judge it
“We will know it when we see it” is honest but costly. It usually means two extra rounds.
Try this instead. Write three words you want someone to use after seeing the work. Then three words you do not want them to use.
We want: confident, clear, grown-up.
We do not want: playful, cheap, corporate.
Six words. Now anyone in the room can point at a design and say whether it does the job, without it turning into taste.
A whole brief, on one page
What: A brand identity and a five-page site.
For: Operations managers at mid-size food producers.
Problem: We look like a startup and we sell to people who buy from established suppliers.
Fixed: The name stays. Live before 1 March.
Good looks like: trusted, precise, calm. Not quirky, cheap or corporate.
Budget: $25,000 to $35,000.
That took ten minutes to write. It would let any studio give you a straight answer, and it would let you compare two answers fairly.
What happens if you skip it
Without a brief, the first round becomes the brief. You look at work, you react, and slowly everyone works out what you meant.
That does work. It just costs two or three extra weeks, and you pay for those weeks.
Worth remembering
In four lines
- One page beats twenty. Long briefs get skimmed.
- Describe the problem, not the solution you assume you need.
- Write down what cannot change. Unwritten constraints cost the most.
- Pick three words you want and three you do not.
Questions we get asked
What if we do not know what we want?
Say that plainly. It is a real and common starting point, and it changes what gets proposed. Pretending to know costs more than admitting you do not.
Should we include competitor examples?
A few, with a line on what you like or dislike about each. Twenty with no notes tells us nothing.
Who should write it?
Whoever can approve the work. If the brief comes from someone who cannot sign off, you will get a second brief later.
Read next
Working on this yourself?
We help teams sort this out for a living. Tell us what you’re building, or see how it went for other people.