Creative Brief
by
Weekly creative briefs for ecommerce teams: read what the account is actually short of, then write the briefs that fill the gaps rather than repeating winners.
Creative briefs usually say make more of what worked, which produces a library of variations on one idea and an account that collapses the moment that idea fatigues. The useful brief comes from the opposite direction: which angles, formats and audiences the account has no coverage for, and which of those gaps is costing the most. That requires looking at the whole library as a portfolio rather than at last week's top performer. This skill reads what the account is actually short of and writes the briefs that fill the gaps.
What It Does
- Maps the existing library by angle, format and audience to find the gaps.
- Ranks the gaps by the spend they would let you deploy.
- Reads which objections your ads never address, from support and reviews.
- Separates a fatigued winner needing a refresh from a genuinely missing angle.
- Specifies the hook, the proof and the format for each brief.
- Generates the week's brief list, ordered by what each unlocks.
Use Cases
- Stop producing the fifth variation of the same concept.
- Find the angle your ads have never tried.
- Brief against the objection your support inbox sees every week.
- Give a creative team specs rather than a request for more.
Example Session
What should the creative team make this week?
Four briefs. The gap that costs most is the one you have never filled.
- Brief | Gap it fills | Evidence | Spend it unlocks | Priority
- Fit and sizing demo | No ad addresses fit | 28% of tickets | $18,000 per month | 1
- Durability, 12 month wear | No longevity proof | Review theme | $11,000 per month | 2
- Unboxing refresh, new hook | Fatigued winner | Frequency 6.4 | Maintains $14,000 | 3
- Static, price-led | No static assets at all | Format gap | $6,000 per month | 4
Nothing in your library addresses fit, and fit generates more than a quarter of your support tickets, which means it is stopping purchases at a much higher rate than it generates questions. That brief is first. The unboxing refresh is third rather than first because it protects existing spend rather than unlocking new spend, which is a real job and a smaller one.
Why This Beats Prompting It Yourself
More Of What Worked vs What The Library Is Missing: The Hidden Risk
A generic prompt briefs variations on the top performer, because the top performer is the visible evidence of what works. Six months of that produces an account with one idea expressed nine ways, which scales until the audience exhausts it and then has nothing to fall back on. The gap is invisible when you only look at winners.
When You Still Need The Objection Nobody Advertised Against
Support tickets and reviews are a list of the reasons people hesitated, written by the people who hesitated. Almost no creative brief reads them. Fit appearing in a quarter of tickets means it is stopping far more purchases than it generates questions, and not one ad in the library addresses it.
Known Limitations
Spend unlocked per gap is an estimate from how comparable angles have performed in your account and in the category, and a genuinely new angle carries a wider band the output labels. Library mapping depends on your creative being taggable by angle and format, which for a large unlabeled library takes a first pass. It writes the briefs and does not produce assets.
Install It
- Download the ZIP.
It is free and there is no account to create. - Unzip it into your agent's skills folder.
Claude Code reads ~/.claude/skills/, which is hidden by default: the command in the folder block opens it. Other agents scan their own directory, so drop the same folder there instead. - Ask your agent to use it.
Restart the agent if it was already running, then it picks the skill up with no config.
skills folder
Copy
~/.claude/skills/creative-brief/
SKILL.md
references/angle-map.md
briefs/
# one brief per asset, with the gap it fills
# macOS: create the folder and open it in Finder
mkdir -p ~/.claude/skills && open ~/.claude/skills
# Windows: paste in the Explorer address bar
%USERPROFILE%\.claude\skills\Frequently asked questions
How is this different from briefing more of the winner?
It reads the library as a portfolio and finds what is not covered, then ranks those gaps by the spend they would let you deploy. Refreshing a winner is still on the list, and it protects existing spend rather than unlocking new spend, which is why it usually ranks lower.
Which agents does the skill run in?
Any agent that supports the open SKILL.md format: Claude Code, Cursor, Codex CLI, GitHub Copilot, Gemini CLI, Manus, Grok and others load it unmodified. The format is portable, the location is not. Each agent scans its own skills directory, so you drop the same folder into whichever one yours uses. There is no config file to edit and no API key to provision.
What does it need connected to work?
Creative-level performance data, your asset library with enough metadata to classify angle and format, plus support tickets and reviews. The tickets are what make the objection analysis possible and are the part most teams have not connected.
Why prioritize an angle that has never been tested?
Because an untested angle addressing a documented objection has more upside than a ninth variation of a proven one, and the evidence for it is already in your support inbox. The skill does not guess at new angles; it briefs against objections you can show are real.
What goes into each brief?
The gap it fills, the hook, the proof point to use, the format and length, and what to avoid. It is specific enough to produce from and stops short of art direction, which is not a thing a skill should be doing.
What is included with the skill?
The skill itself, the angle map reference, the brief format, and the gap ranking model. It is free and security scanned. We re-publish the ZIP when the model changes, so download it again if your priorities start shifting.
A brief is only as good as the performance read behind it.
Polar puts spend, revenue and contribution margin per creative on one definition, so next week's briefs aim at what actually earned.
<script type="application/ld+json">{"@context":"https://schema.org","@type":"FAQPage","mainEntity":[{"@type":"Question","name":"How is this different from briefing more of the winner?","acceptedAnswer":{"@type":"Answer","text":"It reads the library as a portfolio and finds what is not covered, then ranks those gaps by the spend they would let you deploy. Refreshing a winner is still on the list, and it protects existing spend rather than unlocking new spend, which is why it usually ranks lower."}},{"@type":"Question","name":"Which agents does the skill run in?","acceptedAnswer":{"@type":"Answer","text":"Any agent that supports the open SKILL.md format: Claude Code, Cursor, Codex CLI, GitHub Copilot, Gemini CLI, Manus, Grok and others load it unmodified. The format is portable, the location is not. Each agent scans its own skills directory, so you drop the same folder into whichever one yours uses. There is no config file to edit and no API key to provision."}},{"@type":"Question","name":"What does it need connected to work?","acceptedAnswer":{"@type":"Answer","text":"Creative-level performance data, your asset library with enough metadata to classify angle and format, plus support tickets and reviews. The tickets are what make the objection analysis possible and are the part most teams have not connected."}},{"@type":"Question","name":"Why prioritize an angle that has never been tested?","acceptedAnswer":{"@type":"Answer","text":"Because an untested angle addressing a documented objection has more upside than a ninth variation of a proven one, and the evidence for it is already in your support inbox. The skill does not guess at new angles; it briefs against objections you can show are real."}},{"@type":"Question","name":"What goes into each brief?","acceptedAnswer":{"@type":"Answer","text":"The gap it fills, the hook, the proof point to use, the format and length, and what to avoid. It is specific enough to produce from and stops short of art direction, which is not a thing a skill should be doing."}},{"@type":"Question","name":"What is included with the skill?","acceptedAnswer":{"@type":"Answer","text":"The skill itself, the angle map reference, the brief format, and the gap ranking model. It is free and security scanned. We re-publish the ZIP when the model changes, so download it again if your priorities start shifting."}}]}</script>