Creative Strategist
by
Weekly creative strategy for ecommerce teams: read fatigue across every platform at once, then brief next week's assets against what the portfolio is missing.
Creative fatigue is measured per platform, because that is where the frequency numbers live. The audience is not per platform. The same person sees your ad on one feed and then another, so an asset can look fresh in each account and be thoroughly worn out in the only place that matters, which is in front of the customer. Reading fatigue one platform at a time systematically underestimates it. This skill reads fatigue across every platform at once and briefs next week's assets against what the portfolio is actually missing.
What It Does
- Reads frequency and fatigue across platforms rather than inside each one.
- Estimates true exposure where the same audience overlaps two feeds.
- Identifies the concept that is tired everywhere versus tired in one place.
- Maps the portfolio by angle to find what has no coverage.
- Ranks what to produce by the spend each gap would unlock.
- Briefs next week's assets, with the concept to retire and the one to keep.
Use Cases
- Catch the asset that looks fresh per platform and is exhausted overall.
- Stop retiring a concept that is only tired on one feed.
- Brief against the angle the whole portfolio is missing.
- Give the creative team a week's work ordered by what it unlocks.
Example Session
What is tired across our accounts, and what should we make next week?
One concept is worn out everywhere. Another only looks tired on one feed.
- Concept | Freq, platform A | Freq, platform B | True exposure | Verdict
- UGC unboxing | 3.1 | 3.4 | 5.9 | Retire, exhausted
- Founder to camera | 5.8 | 0.4 | 6.0 | Retire, exhausted
- Before and after | 2.2 | 1.1 | 2.9 | Keep, scale
- Problem-solution | 0.9 | 0.6 | 1.4 | Keep, produce more
The unboxing looks acceptable on both platforms at around three and is near six in front of the overlapping audience, which is why it has been declining for two weeks with no per-platform explanation. Problem-solution is the angle to produce against: the portfolio has almost no coverage of it, and it is the one your support inbox says customers need.
Why This Beats Prompting It Yourself
Per-Platform Frequency vs True Exposure: The Hidden Risk
A generic prompt reads each platform's frequency report and finds nothing alarming, because 3.1 and 3.4 are both fine. The customer saw the asset almost six times. Every platform is telling the truth about its own delivery and none of them can see the other, so the decision made from either one is wrong in the same direction.
When You Still Need Tired Everywhere Told Apart From Tired Here
The founder ad is at 5.8 on one platform and 0.4 on the other, which reads as a platform problem and is not: the overlapping audience has seen it six times regardless. Meanwhile a concept genuinely tired in one place and fresh in another should be rotated rather than retired. Only the combined view distinguishes those two.
Known Limitations
True exposure is estimated from audience overlap between platforms, which is inferred rather than measured because no platform exposes the other's delivery, and the output states the confidence. Portfolio mapping needs creative taggable by angle. It briefs the assets and does not produce them.
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-strategist/
SKILL.md
references/exposure-model.md
weekly/
# one file per week, with the briefs and retirements
# 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 checking frequency in each platform?
Each platform reports only its own delivery, so an asset running on two feeds looks half as worn as it is. This estimates exposure across the overlapping audience, which is the number that actually predicts when performance falls off.
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 delivery and performance from each platform, audience definitions, and your asset library with angle metadata. Support tickets and reviews feed the gap analysis and are optional.
How accurate is the true exposure estimate?
It is an estimate with a stated confidence, built from audience overlap and delivery patterns. It is considerably closer than any single platform's number, and the skill is explicit that it is inferred rather than measured.
Should a tired concept always be retired?
No. A concept tired on one platform and fresh on another gets rotated rather than retired, and a genuinely strong concept that is exhausted gets refreshed with a new execution rather than replaced. Retirement is for concepts that are worn out everywhere and were never strong.
What is included with the skill?
The skill itself, the exposure model reference, the weekly brief format, and the portfolio mapping rules. It is free and security scanned. We re-publish the ZIP when the model changes, so download it again if your readings start shifting.
Creative fatigue is visible only across platforms at once.
Polar reconciles spend and net revenue per asset across 45+ sources, so next week's brief is built on one performance read rather than three.
<script type="application/ld+json">{"@context":"https://schema.org","@type":"FAQPage","mainEntity":[{"@type":"Question","name":"How is this different from checking frequency in each platform?","acceptedAnswer":{"@type":"Answer","text":"Each platform reports only its own delivery, so an asset running on two feeds looks half as worn as it is. This estimates exposure across the overlapping audience, which is the number that actually predicts when performance falls off."}},{"@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 delivery and performance from each platform, audience definitions, and your asset library with angle metadata. Support tickets and reviews feed the gap analysis and are optional."}},{"@type":"Question","name":"How accurate is the true exposure estimate?","acceptedAnswer":{"@type":"Answer","text":"It is an estimate with a stated confidence, built from audience overlap and delivery patterns. It is considerably closer than any single platform's number, and the skill is explicit that it is inferred rather than measured."}},{"@type":"Question","name":"Should a tired concept always be retired?","acceptedAnswer":{"@type":"Answer","text":"No. A concept tired on one platform and fresh on another gets rotated rather than retired, and a genuinely strong concept that is exhausted gets refreshed with a new execution rather than replaced. Retirement is for concepts that are worn out everywhere and were never strong."}},{"@type":"Question","name":"What is included with the skill?","acceptedAnswer":{"@type":"Answer","text":"The skill itself, the exposure model reference, the weekly brief format, and the portfolio mapping rules. It is free and security scanned. We re-publish the ZIP when the model changes, so download it again if your readings start shifting."}}]}</script>