How we write briefs that don't lie

Most bad design work was doomed before anyone opened a file, because the brief was lying. Not maliciously — briefs lie the way holiday plans lie, by describing a world where nothing conflicts. Every stakeholder's wish is included, every goal is a priority, the audience is "everyone, but especially young professionals," and success will somehow be measured later.

Then the work arrives, reality casts its vote, and the review meeting becomes the place where the brief's contradictions finally get resolved — expensively, in front of the finished work, at the moment when changing course costs the most.

A brief that doesn't lie is a document of decisions, not intentions. Over the years we've reduced this to four tests. A brief that passes them is short, slightly uncomfortable to write, and almost boringly effective.

Test one: it names what we're not doing. Anyone can write goals. The honest brief also carries a "not doing" list: not redesigning the logo, not chasing the enterprise segment this quarter, not optimizing for awards. Scope is only real when its edges are written down — and every fight you'll otherwise have in week six lives on that unwritten list.
Test two: priorities have an order, not a tie. "Fast, premium, and accessible" isn't a priority list; it's a decorative sentence. The lying stops when the brief says what wins in a conflict: when premium feel and page speed collide, speed wins. If the stakeholders can't agree on that sentence before the work starts, the brief's job is to force that argument now — while it's cheap — instead of hiding it until the review.

Test three: success has a number and a date. Not because design reduces to metrics, but because "we'll know it when we see it" is a blank check written against the designer's account. Even a rough measure — trial signups up 15% by Q3, support tickets about checkout down by half — changes the entire conversation from taste to purpose. If nobody can name a number, that's a finding: the project's purpose isn't understood yet, and no amount of design will fix that.

Test four: it records who decides. Feedback comes from everywhere; decisions must come from somewhere. One named person who resolves conflicts, written into the brief. Every project that died by committee died here, and every version of "we'll align as a group" is this test failing politely.

Notice what these four have in common: each one relocates a conflict from the end of the project to the beginning. That's the entire technology. A brief isn't a form to fill — it's a negotiation you conduct early, on purpose, while disagreement is still cheap. The discomfort you feel writing an honest brief is the exact discomfort you're removing from the review meeting.

Which is why our rule of thumb is this: if the brief was easy to write, it's probably lying. Send it back and ask what it's hiding.