We audited 27 public submissions to a $120,000 clipping campaign. 15 fail on caption rules alone.
Call of Duty's Modern Warfare 4 reveal ran one of the biggest clipping campaigns of the summer: a $120,000 pool, 11,574 creators, 94.1 million views, fully paid out. The campaign brief is public, and it's a good one: clear disclosure rules, tagging rules, a hashtag cap, length and language requirements. The operator reported a 56% approval rate, which means someone had to review thousands of submissions and reject nearly half.
We pulled 27 public TikTok submissions to the campaign and graded them against the brief's own rules, using only what's visible in the post itself: no judgment calls and no video review yet, just the objective checklist a reviewer has to apply to every submission.
The results
12 of 27 pass. 15 fail at least one rule. The failure profile:
Most failing posts break more than one rule, and the individual failures are the kind that slip through manual review:
- A submission with 79,400 views carries the full campaign tag block but a caption in another language. The brief says English only. That's real money paid against a rule the post visibly breaks.
- A submission with 102,400 views has no disclosure hashtag at all. The brief says missing disclosure means automatic rejection, no exceptions.
- One account posted eight near-identical clips with copy-pasted example captions from the brief. The brief bans exactly this.
- The most common failure is placement. The disclosure exists but sits mid-caption or stacked with other tags, where the brief requires it first and alone. Ten posts get this wrong, and it is the kind of miss a skimming reviewer makes, because "#ad is in there somewhere" reads as compliant at a glance. FTC disclosure guidance is specific that placement matters.
Why this matters if you run campaigns
That 56% approval rate means the operator's reviewers did real work, and still, over half of the surviving public posts fail at least one objective caption rule. This is not a criticism of any operator. Reviewing submissions from 11,574 creators is triage, and under triage the first rules lost are placement and formatting rules, which are precisely the ones disclosure regulators care about. Every miss is either money paid out against the brief or compliance exposure for the brand.
One more thing we found: briefs contradict each other. This campaign mandates #Ad as the first element of every caption, while another live campaign we reviewed the same week bans any sponsored-looking language in captions outright. A reviewer working both campaigns has to apply opposite rule sets post by post without slipping. Holding a distinct checklist per campaign, consistently, across thousands of posts is exactly the work we built valid8 to automate.
Method and limitations
We used the public brief (linked from the campaign's public listing), public posts, and caption and metadata rules only. The language check is script-based, and disclosure placement is graded as "first hashtag in the caption" because line-break data is imperfect. Video-level rules (footage source, audio, meme edits) need frame review, which is what the full valid8 audit does; video review can only add failures to these numbers, not remove them. Posts are anonymized by design, since the point is the pattern, not any individual clipper.
valid8 grades every clip in a campaign batch against the campaign's own brief: pass or fail, the reason per rule, the exact fix. Two AI reviewers from different model families, stricter verdict wins, an audit trail you can forward to the brand. Book a pilot or write to contact@i8l.tech.