How to vet creators before outreach: an evidence-led workflow

Creator vetting is not a hunt for one magic score. It is a documented comparison of the audience you need, the content a creator actually publishes, the response that content earns, and the risks a partnership would introduce.

Start with a comparable method

Engagement rate can flag unusual accounts, but a universal cutoff is misleading. Calculate it consistently, state the post window, and compare creators within the same platform, format, audience-size band, and market. A Reel, a long-form video, and a static post do not create equivalent response opportunities.

Use the calculator as a screening aid, then inspect the distribution behind the average. A stable pattern across recent posts is more informative than one viral outlier. The 2024 meta-analysis also reports mixed results for follower count and shows that post, follower, influencer, platform, and product context all affect outcomes.

Collect independent evidence

• Audience match: record the required countries, languages, and interests before opening profiles; mark unavailable data as unknown rather than favorable.
• Comment substance: sample recent comments for topic-specific questions and replies. Repeated generic reactions are a review flag, not proof of fraud.
• Content and brand fit: review recent themes, production consistency, competitor conflicts, claims, and the density of paid partnerships.
• Disclosure readiness: the FTC says material connections include payment and free or discounted products, and disclosures should be hard to miss. Meta also treats gifted products as an exchange of value for branded-content labeling.
• Suspicious influence signals: look for abrupt unexplained growth, inactive-looking audiences, and response patterns that do not fit the content. The FTC defines fake indicators narrowly; do not label an account fraudulent from a heuristic alone.

Turn evidence into a decision

Define the decision rule before reviewing names. That prevents a famous profile or a high headline rate from quietly changing the standard. Keep three outcomes so missing evidence is not confused with failure.

1. Qualify when required audience and content-fit evidence is present, no unresolved hard-stop risk exists, and comparable performance is credible.
2. Manual review when a required field is unavailable, a pattern is unusual, or disclosure and usage-right expectations need confirmation.
3. Decline when a documented hard stop is met, such as market mismatch, direct conflict, unsafe claims, or unwillingness to meet disclosure requirements.
4. Record the evidence date and reviewer so the decision can be refreshed instead of treated as permanent truth.

Sources and review notes

1. Influencer marketing effectiveness: A meta-analytic review — Journal of the Academy of Marketing Science
2. Disclosures 101 for Social Media Influencers — U.S. Federal Trade Commission
3. The Consumer Reviews and Testimonials Rule: Questions and Answers — U.S. Federal Trade Commission
4. What is considered branded content — Instagram Help Centre