A client once arrived with a brief for a website redesign. Sixteen pages, a competitor audit, a moodboard, a budget. What they needed was to stop selling to procurement teams. The website was fine. The website was never the problem.
This is not a failure of the client. Briefs are written in the language of things a company knows how to buy, and no one has ever been able to buy "our positioning is confusing." So it gets translated into the nearest purchasable object, which is usually a website.
Ask what changed
The most useful question in a kickoff is not "what do you want" but "what changed." Something always did. A competitor raised. A founder left. A metric that used to go up stopped. The brief is the symptom; the change is the cause.
If nothing changed, nobody would be paying an agency. The change is the brief.
Separate the observation from the prescription
Clients are extremely good at observation and, understandably, less practised at prescription. "Our homepage is confusing" is usually a reliable observation. "So we need an explainer video" is a prescription, and prescriptions are ours to earn.
We read every brief twice: once to collect what the client has noticed, and once to strike out what they have proposed. The first list is nearly always accurate. The second is a starting position, not a specification.
Say it back before you solve it
Before any work begins we return a one-page restatement of the problem. Not a proposal — a restatement. If the client reads it and says "yes, that is it, and we had not put it that way," the engagement will go well. If they argue with it, that argument is far cheaper now than it will be in week nine.
- What changed, in one sentence
- What that is costing, in a number if one exists
- What we believe is actually causing it
- What we would have to be wrong about for this to fail
That last line is the one people remember. Naming your own failure condition in week one buys you an enormous amount of trust in week ten, when something inevitably needs to change.