I spend most of my week in the gap between what a client had in their head and what turns up in their inbox. When those two things match, the project feels effortless and everyone thinks the designer is a genius. When they don’t, we’re three rounds deep, the deadline has moved, and nobody is enjoying themselves.
The gap almost always starts at the brief. Not because the client wrote a bad one. Most people have never been taught to write one at all, and there’s no reason they should have been. Briefing is an agency skill that clients are quietly expected to arrive with.
So here’s the version I’d write if I were sitting on your side of the table.
Revision rounds aren’t the problem
Every quote we write includes two revision rounds, which is plenty for a normal job. The projects that blow past them rarely do it because the design was wrong. They do it because the design answered a question the business hadn’t settled yet.
Who is this for. Where does it run. What does it need to make someone do. If those are still open when work starts, the first proof becomes the meeting where you decide them, except now you’re deciding them by reacting to artwork instead of talking it through. That’s an expensive way to make up your mind, and it’s the single most common reason a design project runs long.
A brief is a decision record, not a wish list
The useful way to think about a brief is as a written record of what’s already been agreed, so nobody re-litigates it at proof stage.
The best brief I’ve been sent was four sentences and a date. The worst ran eleven pages and I couldn’t have told you what the thing was for. Length is not the variable. Clarity about the job is.
The one page that prevents most of it
Five lines, and you can write them in ten minutes.
What it physically is. An A4 flyer, six paid social variants, a 24-slide deck. Be literal, because “some assets for the campaign” turns into a scoping call.
Where it runs, and to what spec. Channel, placement, dimensions, print or screen. Someone has to get the specs from the media buyer or the printer, and if it isn’t you it’ll be us asking you to ask them, which costs two days.
Who it’s for. One sentence describing an actual person, and what they already know about you. “Existing customers who have bought once” and “people who’ve never heard of us” produce different work.
What it has to make them do. Describe the behaviour you’re after. If you find yourself writing that it should feel premium, keep going until you reach what somebody actually does about that.
When it’s due, and what happens then. A real date, and what it’s tied to. We schedule differently for a media booking that can’t move than for an internal review.
Five questions worth answering before you send it
What has to be true for this to have worked? If you can’t answer that, the work has no target and every version is arguable.
What’s already fixed? Brand guidelines, legal wording, mandatory logos, the fact that your MD won’t approve anything with orange in it. Tell us the constraints up front. We’d rather design inside them than discover them at round two.
Is the copy final? If it isn’t, say so. Designing around placeholder text and then dropping in forty per cent more words is not a tweak, it’s a redesign, and it will use a revision round you’d rather spend on something else.
Who signs it off, and are they around? The most common cause of a late project in my experience isn’t design time. It’s an approver on leave in the week we deliver. Name the approver, check the calendar, and if there are five of them, decide now who breaks a tie.
What have you already tried and rejected? Send it. Nothing saves a round faster than knowing which direction is already dead.
What to send with it
References help, but only with a sentence attached. “This one, for the way the type sits over the image” tells us something we can use. “This one” tells us you like something about it and leaves us guessing which part.
Send your files too. Logos in vector, the fonts you’re licensed for, the guidelines PDF even if it’s three years old and half-ignored. We read guidelines before opening an artboard, and where they’re silent we make a call and write it down so your system gets stronger rather than drifting.
How to give feedback that gets you closer
Describe the problem, not the fix. “It feels too corporate for this audience” is more useful than “make the logo bigger”, because the second one is a solution to a problem you haven’t told us about, and there might be a better solution to it.
Consolidate. One thread, one set of comments, one person collating. Five stakeholders emailing separately guarantees contradictions that somebody then has to referee, usually me, usually on a Friday.
And say what’s working, not to be nice, but so we don’t accidentally take it away in the next version. Half the feedback rounds I’ve watched go sideways went sideways because a designer removed the one element everybody liked and nobody had mentioned.
What you should get back
Questions, first. If an agency takes your brief and asks nothing, either it was an exceptional brief or nobody read it properly.
A quote with the revision rounds written into it, so the scope conversation happens before there’s tension rather than after. Proofs that come annotated, with the reasoning explained rather than just presented. Artwork exported to each channel’s real spec, correct colour profiles and all. And the packaged source files, handed over as standard, because you paid for the artwork and you should own it.
You should also know who is actually doing the work. On our design jobs Mark Hebblewhite works the brief with you directly rather than through an account-manager relay, which removes an entire layer where meaning goes missing.
If you only do one thing
Write down what the work has to make someone do, and put it at the top of the page. Everything underneath it is negotiable, and everything gets easier once that line exists.
The bigger the job, the more it pays off. When Revolve Partners came to us there was no digital footprint at all, so the brief had to carry an entire identity and the website that followed. That project ran well because the questions above got answered early, not because anybody guessed correctly.
If you want a second pair of eyes on a brief before it goes anywhere, send it through. I’ll tell you what’s missing, and there’s no charge for that. Our creative team would rather spend the time at the start than at round three, and so, I promise, would you.