Briefs 5 min read

The one-line brief: how to ask for marketing work

Most briefs fail before the work starts — too vague to act on, or so detailed they're the work itself. Here's the anatomy of a single sentence that a marketer, human or AI, can run with.

Every piece of marketing starts as a request. “Can you make some noise about the launch?” “We should be doing more on LinkedIn.” “Get us more signups.” The work that follows is rarely better than the sentence that asked for it — and most of those sentences are terrible.

We’ve read thousands of briefs since building Spellsuite, because a one-line brief is literally our product’s input. The good ones share a shape. It fits in one sentence, and it has three parts.

The anatomy: outcome, audience, constraint

A brief that works names what should happen, for whom, and within what limit. That’s it. Compare:

"do some marketing for the pottery studio"
"post 3x/week on IG, run CHF 20/day search ads on
  'töpferkurs zürich', A/B two landing pages, and…"
"fill the September beginner course with Zürich
  locals, under CHF 800 total spend"

The first gives the executor nothing to aim at. The second isn’t a brief at all — it’s a to-do list that smuggles in every decision, so the executor can’t apply any judgment (and can’t be blamed for the outcome, either). The third states a goal you can verify, an audience you can find, and a budget you can’t blow. Everything else is legitimately the executor’s job.

Why vague briefs feel safe (and aren’t)

Vagueness feels like flexibility: “I don’t want to constrain the creative.” In practice it just moves the decisions downstream, to whoever executes — who then guesses, and gets corrected in review, three days later. The brief didn’t avoid the decisions; it deferred them to the most expensive possible moment.

Rule of thumbIf two reasonable people could read your brief and produce work for two different audiences, the brief is missing its audience clause. Same for outcome and constraint.

Outcomes, not deliverables

“Write four blog posts” is a deliverable. “Be the answer when someone asks an assistant about beginner pottery in Zürich” is an outcome. Deliverable briefs get you exactly what you asked for, which is the problem: if the four posts don’t move anything, the brief was still technically satisfied.

Outcome briefs let the executor choose the deliverables — maybe it’s two posts, a rewritten landing page, and fixing your Google Business listing. When the executor is an AI suite that can draft all of those in one pass, this stops being a philosophical point and becomes the entire value: the brief is the steering wheel, and deliverable-briefs weld it in place.

The constraint is the trust

The constraint clause — budget cap, deadline, brand line you won’t cross — is what makes delegation safe. It’s the difference between “go do things with my ad account” and “go do things with my ad account, capped at CHF 40 a day.” Nobody delegates without it, and no one should:

  • Money: a daily or total cap, always. “Reasonable spend” is not a number.
  • Time: when it needs to be live, and when to stop trying.
  • Voice: the one thing you’d never say. One is usually enough.

A good executor treats constraints as hard walls, and it’s exactly because the walls are hard that you can hand over everything inside them.

Try it on your own backlog

Take the top item on your marketing to-do list and rewrite it as outcome + audience + constraint, one sentence. If you can’t — if you don’t know the outcome, or can’t name the audience, or won’t commit to a cap — you’ve found the real blocker, and it was never the execution.

One line is the whole job description.

Type a brief like this into Spellsuite and watch it become ads, pages, and a plan.

Start free
Written by the Spellsuite team · The Spellbook All posts