Definition
A good creative brief is one that passes 7 specific criteria - a measurable goal, behavioral audience, sharp angle, named hook archetype, explicit do-nots, real references, and a passing 5-minute test. Briefs that meet all 7 produce shippable creative on the first cycle; briefs that miss any 1 generate rewrites.
Why it matters
What this unlocks
- 1
Most teams don't evaluate briefs before sending them - they evaluate creative after. By the time the creative comes back wrong, the brief has already cost a week and a budget.
- 2
The 7 criteria are diagnostic: if a brief fails any one, the fix is targeted. You don't have to rewrite the whole document - just the failing section.
- 3
Briefs that pass all 7 criteria deliver 3-5x faster from draft to shippable creative. The criteria are how senior teams compress timeline.
- 4
AI generators (like Shuttergen) perform much better on criteria-passing briefs. Generic briefs produce generic AI output; specific briefs produce specific output. The brief is the bottleneck.
Parts
What's inside
Criterion 1: Measurable goal with a number
One business outcome, one target number, one timeframe. 'Drive trials at $35 CAC over 60 days'. If the goal can't fit in one sentence with a number, the upstream strategy isn't decided yet - fix that before writing the brief.
Criterion 2: Behavioral audience
Two behaviors + one tension. 'Endurance athletes who train 5+ hours weekly AND use generic electrolytes they're frustrated with'. Demographics describe; behaviors predict. The audience line is where most briefs collapse into generic.
Criterion 3: Sharp angle (one frame on the product)
Not a feature list. One sentence about what the ad is *about*, not what the product *does*. Sharp enough that a competitor wouldn't say it. If your angle could appear in a competitor's brief without modification, it's not sharp enough.
Criterion 4: Named hook archetype
Pick from the standard set: problem→solution, transformation, day-in-the-life, founder-to-camera, demo, testimonial, comparison. Adjectives ('engaging') aren't archetypes. The named archetype tells the receiver the structural shape so they can focus on the specific content.
Criterion 5: Explicit do-nots (3-5)
What the ad must NOT do. Each do-not should represent specific category knowledge or campaign learning. Generic do-nots ('don't be too sales-y') don't count. The do-nots are where distinctive creative lives.
Criterion 6: Real references (5-10 links)
Span direct competitors (benchmark), adjacent categories (differentiation), and organic content (tone). Skip the prose description if the links do the job. References compress 500 words of description into 5 minutes of skimming.
Criterion 7: Passes the 5-minute test
Hand the brief to someone not on the campaign. They should be able to (1) describe the ad in one sentence and (2) name three things the ad must NOT do, in under 60 seconds. If they hesitate or fish for context, the brief isn't done.
Shuttergen
Auto-check your brief against 7 criteria - in 10 seconds.
Shuttergen runs the 7-criteria check on every brief you write and flags fails with specific fixes. Catch brief-quality issues before they cost a week of creative cycles.
Worked example
Running the 7 criteria against a real brief
The brief: a DTC hydration brand's Q3 acquisition brief, written by the marketing lead in 90 minutes.
Criterion 1 (measurable goal): 'Drive 1,000 new-customer trials of the 30-day starter at $29 AOV within 60 days, at $35 max CAC.' One outcome, three numbers, one timeframe. Pass.
Criterion 2 (behavioral audience): 'Endurance athletes who train 5+ hours per week AND currently use a generic or store-brand electrolyte they're frustrated with.' Two behaviors, one tension. Pass.
Criterion 3 (sharp angle): 'Higher sodium ratio than mainstream alternatives, third-party tested for purity, no artificial dyes.' Three testable claims, tied to the audience's frustration. Sharp - a competitor making generic electrolyte wouldn't say it. Pass.
Criterion 4 (named hook archetype): 'Problem→solution. Open on dehydration symptoms during a long ride; transition to product use; end on next-ride performance contrast.' Named, structural, specific. Pass.
Criterion 5 (do-nots): '(1) No price-led hook. (2) No bottle-shot static. (3) No founder-to-camera. (4) No 'pro athlete' framing.' Four do-nots, each tied to specific campaign learning. Pass.
Criterion 6 (references): 5 ads from competitor X, 3 from adjacent supplement brand Y, 2 from a running content creator. Pass.
Criterion 7 (5-minute test): Handed the brief to the head of CX. In 45 seconds she said: 'It's an ad about how this electrolyte has more sodium than generics, opening on cramping during a long ride - and we won't lead with price, won't use static, and won't have the founder on camera.' Pass.
Outcome: brief passed all 7 criteria. Shipped to the editor on Monday; first cuts back Wednesday; final approval Thursday. 4 days from brief to live creative. Briefs that fail criteria typically take 10-14 days because of rewrite cycles.
Common mistakes
What people get wrong
Treating the criteria as nice-to-haves
Each criterion has a specific cost when missed - vague goal produces hedged creative, demographic audience produces generic ads, missing references produces visual confusion. The cost is real every time; the criteria aren't optional.
Evaluating after the brief ships
Run the 7 criteria before sending. Catching a missing do-not at brief time costs 5 minutes; catching it after the creative comes back wrong costs a week. The criteria are diagnostic; use them upstream.
Conflating 'long' with 'good'
Briefs that fail the criteria don't get better by adding more sections. Length is uncorrelated with quality. A 1-page brief that passes all 7 criteria beats a 6-page brief that fails 3 of them.
Substituting templates for decisions
Filling a template with empty placeholders or generic descriptions doesn't pass the criteria. The criteria measure decisions, not formatting. Templates accelerate documentation, not strategy.
Skipping the 5-minute test
The 5-minute test is the cheapest quality check and the most reliable. Teams that skip it ship briefs that fail at the criteria stage 40% of the time. 5 minutes vs a week of rewrite cycles - always run the test.
Why the 7 criteria exist in this specific order
Goal first because everything downstream depends on it. Hedged goals produce hedged audiences, hedged angles, hedged hooks. The cascade starts at the top. If criterion 1 fails, fixing criteria 2-7 won't save the brief.
Audience second because angle depends on it. You can't write a sharp angle without knowing who you're sharpening it for. Behavioral audience first, then angle. Teams that write angle before audience produce briefs that are technically interesting but not strategically grounded.
Angle and hook archetype together because they're a pair. The angle is *what* the ad is about; the hook archetype is the *structural shape* that delivers it. A great angle inside the wrong archetype underperforms; a great archetype with a vague angle produces forgettable work.
Do-nots and references because they protect the downstream execution. Do-nots prevent off-strategy moves; references anchor visual language. Both prevent the receiver from inventing things that contradict the brief.
5-minute test last because it validates the whole. The first 6 criteria are component-level; the 5-minute test measures whether the components add up. Briefs can pass all 6 components and still fail the integration test - which is when you know to cut, not add.
What happens when each criterion fails - and the fix
Goal fails (vague or hedged): creative comes back covering too many themes. Fix: pick ONE outcome, kill the others or move them to a separate brief.
Audience fails (demographic): creative reads as generic - could fit any brand in the category. Fix: rewrite as two behaviors + one tension. Reference customer reviews or sales calls for the behavioral language.
Angle fails (feature list): creative looks like a product page. Fix: pick one frame on the product - sharp enough that a competitor wouldn't say it. Kill the others.
Hook archetype fails (adjective instead of structure): creative comes back with unclear opening pattern. Fix: name the archetype from the standard set. Tell the receiver the structural shape.
Do-nots fail (missing or generic): creative defaults to category average. Fix: list 3-5 specific things the ad must not do - each tied to actual category knowledge or campaign learning.
References fail (missing or wrong): creative invents visual language inconsistent with brand. Fix: link 5-10 references across competitors, adjacent categories, and organic content. Skip the prose if the links do the job.
5-minute test fails: brief reader can't describe the ad clearly. Fix: cut, don't add. The signal-to-noise ratio is too low - find and delete the sentences that aren't load-bearing.
Auto-check your brief against 7 criteria - in 10 seconds. Shuttergen runs the 7-criteria check on every brief you write and flags fails with specific fixes. Catch brief-quality issues before they cost a week of creative cycles.
How the 7 criteria differ for AI generators
The same 7 criteria apply, but their weight shifts. AI generators are more sensitive to do-nots and references than human editors are; they're equally sensitive to goal and audience.
Do-nots are more important for AI. Humans bring tacit knowledge - they know your brand voice, they've seen your past work, they pattern-match to what 'feels right'. AI doesn't. The do-nots have to be explicit because there's no tribal knowledge filling in.
References are essential for AI, optional for senior humans. A senior editor can produce shippable work from a brief without references because they have years of category exposure. AI doesn't have that exposure - it needs references to anchor visual language.
Hook archetypes are essential for AI. Named archetypes map cleanly to AI prompt structure. Adjective-based hooks ('engaging') don't. AI produces materially better output when the archetype is named.
The 5-minute test still applies. If a human reader can't describe the ad clearly from your brief, AI won't either. Run the test before feeding the brief to any generator.
Internal: creative-brief for the interactive builder; what-makes-a-good-creative-brief for the properties angle; creative-brief-questions for the Q&A approach.
The 7-criteria checklist as a team ritual
Teams that compound on creative quality run the 7-criteria check at brief sign-off - every brief, every time. The ritual is cheap (10 minutes) and catches 90% of brief-quality issues before they cost a week of creative cycles.
The pattern: the brief author drafts; a second person runs the 7 criteria and flags fails; the author fixes; the brief ships. Two-person review, structured criteria, no rewrites required if both pass.
Teams without this ritual ship 3-4x more rewrites. Each rewrite costs 2-4 days of creative time. Over a year, the unstructured-review team spends 60-80 extra days on rewrites that the structured-review team avoids. That's a full quarter of creative throughput lost to skipped reviews.
FAQ
Frequently asked
What is a good creative brief?
How do I know if my creative brief is good?
What's the difference between 'what is a good brief' and 'what makes a good brief'?
Are all 7 criteria equally important?
How long does it take to evaluate a brief against the 7 criteria?
Can AI evaluate my brief against the 7 criteria?
Related
Keep reading
Auto-check your brief against 7 criteria - in 10 seconds.
Shuttergen runs the 7-criteria check on every brief you write and flags fails with specific fixes. Catch brief-quality issues before they cost a week of creative cycles.