TikTok Ads Specialist
by
TikTok campaign review for ecommerce teams: reconcile TikTok spend to net revenue and blended CAC, then write scale, refresh or reallocate per campaign into Notion.
TikTok reports conversions on its own attribution, and on a platform where creative burns out in days rather than weeks the weekly decision is really two decisions at once: is this campaign working, and is the creative still working. Those get conflated. A campaign is cut when the creative was simply tired, or a creative is refreshed when the campaign was never profitable to begin with. This skill reconciles TikTok spend to net revenue and blended CAC on your own definition, separates campaign performance from creative fatigue, and writes scale, refresh or reallocate per campaign into Notion.
What It Does
- Pulls TikTok spend, delivery and platform-attributed results per campaign.
- Reconciles them to net revenue and blended CAC on your own definition.
- Tracks frequency and performance decay per creative inside each campaign.
- Separates a failing campaign from a campaign with a tired creative.
- Flags the creatives worth reshooting against the ones worth retiring.
- Writes scale, refresh or reallocate into Notion with the reasoning attached.
Use Cases
- Stop cutting a profitable campaign because its lead creative burned out.
- Catch creative decay in week one rather than after the CAC moves.
- Give a creative team a brief list grounded in what actually declined.
- Reconcile TikTok's reported results to the revenue finance recognizes.
Example Session
Review this week's TikTok campaigns and tell me what to scale, refresh or reallocate.
Six campaigns ran. Two need creative, one needs cutting, and they are not the same ones.
- Campaign | Spend | TikTok ROAS | Blended CAC | Creative age | Decision
- Spark-UGC-Set-A | $6,200 | 2.1 | $37 | 9 days | Refresh creative
- Broad-Prospecting | $4,800 | 1.6 | $44 | 24 days | Refresh creative
- Retarget-7d | $1,900 | 3.9 | $68 | 31 days | Reallocate
- Product-Demo-New | $2,400 | 1.8 | $33 | 4 days | Scale, plus 20%
The first two look like campaign problems and are creative problems: both held their CAC until frequency crossed three, then declined in step with it. Brief new assets and leave the targeting alone. Retarget-7d is the opposite case, a good reported ROAS on a $68 blended CAC, which is a campaign problem no new creative will fix.
Why This Beats Prompting It Yourself
Platform ROAS vs Creative Decay: The Hidden Risk
A generic prompt reads TikTok's reported numbers and ranks campaigns. On a platform where creative fatigue moves performance faster than anything else, that ranking blames the campaign for what the asset did. The cut that follows throws away targeting that was working, and the same decline starts again on the next campaign with the same creative.
When You Still Need Fatigue Separated From Performance
Two different problems need two different fixes. This skill states a $37 blended CAC declining in step with frequency at nine days, which is a brief, next to a $68 CAC on a campaign that never worked, which is a cut. Each decision goes to Notion with its reasoning, so the creative team gets a list grounded in a measured decline.
Known Limitations
Blended CAC needs an order source alongside the TikTok account, and without one the skill reports platform figures only and labels them. Creative-level attribution depends on TikTok exposing performance per asset, which it does for most campaign types and not all, so some campaigns are analyzed at campaign level and say so. It writes decisions to Notion and does not change budgets in TikTok.
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/tiktok-ads-specialist/
SKILL.md
references/fatigue-model.md
reviews/
# one review per week, pushed to Notion
# 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 TikTok Ads Manager?
Ads Manager ranks campaigns on TikTok's attribution and does not separate creative fatigue from campaign performance. This reconciles spend to your own blended CAC and tracks decay per asset, which is what decides between a refresh and a cut.
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.
How does it detect creative fatigue?
By tracking performance against frequency per asset rather than by age alone. A creative at nine days with frequency past three is tired; the same creative at nine days and low frequency is not. Age on its own is the least reliable signal on this platform.
What does it need connected to work?
The TikTok ad account and an order source from your store. With both it reports blended CAC. With TikTok alone it still separates fatigue from performance and labels every revenue figure as platform-reported.
Does it write the creative brief?
It names the assets that declined, what declined about them, and at what frequency, which is the input a brief needs. A person writes the brief. The useful part here is knowing which of your creatives actually wore out rather than guessing from the campaign average.
What is included with the skill?
The skill itself, the fatigue model reference, the weekly review template, and the Notion decision format. It is free and security scanned. We re-publish the ZIP when TikTok changes its reporting, so download it again if a review starts missing creatives.
TikTok reports its own version of a conversion.
Polar reconciles TikTok spend to net revenue and blended CAC on one definition, so scale, refresh or reallocate holds up outside the platform.
<script type="application/ld+json">{"@context":"https://schema.org","@type":"FAQPage","mainEntity":[{"@type":"Question","name":"How is this different from TikTok Ads Manager?","acceptedAnswer":{"@type":"Answer","text":"Ads Manager ranks campaigns on TikTok's attribution and does not separate creative fatigue from campaign performance. This reconciles spend to your own blended CAC and tracks decay per asset, which is what decides between a refresh and a cut."}},{"@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":"How does it detect creative fatigue?","acceptedAnswer":{"@type":"Answer","text":"By tracking performance against frequency per asset rather than by age alone. A creative at nine days with frequency past three is tired; the same creative at nine days and low frequency is not. Age on its own is the least reliable signal on this platform."}},{"@type":"Question","name":"What does it need connected to work?","acceptedAnswer":{"@type":"Answer","text":"The TikTok ad account and an order source from your store. With both it reports blended CAC. With TikTok alone it still separates fatigue from performance and labels every revenue figure as platform-reported."}},{"@type":"Question","name":"Does it write the creative brief?","acceptedAnswer":{"@type":"Answer","text":"It names the assets that declined, what declined about them, and at what frequency, which is the input a brief needs. A person writes the brief. The useful part here is knowing which of your creatives actually wore out rather than guessing from the campaign average."}},{"@type":"Question","name":"What is included with the skill?","acceptedAnswer":{"@type":"Answer","text":"The skill itself, the fatigue model reference, the weekly review template, and the Notion decision format. It is free and security scanned. We re-publish the ZIP when TikTok changes its reporting, so download it again if a review starts missing creatives."}}]}</script>