How to Tell if an Influencer Has Fake Followers Before You Pay
Influencer fraud is a structural tax on marketing budgets, and it lands on whoever pays for an audience that is not there. A study by CHEQ and University of Baltimore economist Roberto Cavazos put the cost to advertisers at roughly 1.3 billion dollars a year, about 15 percent of influencer spend at the time, and the FTC now treats both fake followers and undisclosed paid endorsements as deception.[1][2][3] The good news for a brand is that almost everything you need to vet a creator is public, and you can do it before a dollar moves. This is the diligence, from the 60-second check to a full audit.
Key points
- Influencer fraud cost advertisers ~$1.3B/yr, ~15% of spend.[1]
- Engagement rate is the first and fastest authenticity tell.
- Under ~1% on a large account is a warning; under 0.5% is a red flag.
- Generic comments from empty accounts signal a fake audience.
- Public-account vetting is lawful, unnotified, and doable before you pay.[4]
Why this is a budget problem, not a vibe
The scale is documented. The CHEQ and University of Baltimore analysis estimated influencer fraud at around 1.3 billion dollars annually, roughly 15 percent of influencer marketing spend at the time, with mega-influencer posts losing tens of thousands per post to fraudulent metrics.[1][2] That is not a rounding error; it is a systematic transfer from brands to whoever sells inflated audiences.
It compounds with a compliance dimension. The FTC's Endorsement Guides require creators to disclose material connections to brands, and the agency has brought actions against fake reviews and fake-influence sellers, so a creator who misrepresents their audience is adjacent to conduct regulators already police.[3] Vetting is both a financial and a legal safeguard.
The 60-second manual check
Open the creator's recent posts and compare interactions to follower count. A 200,000-follower account averaging 300 likes is posting roughly 0.15 percent engagement, far below healthy for any tier, which is an immediate flag.
Then read the comments rather than counting them. Authentic audiences say specific things tied to the content; fake ones leave generic praise and emoji from accounts with no posts of their own. A large following over a hollow comment section is one of the most reliable quick tells, and it costs you a minute.
Read the growth curve
Organic accounts grow along a relatively smooth curve punctuated by occasional spikes from a viral moment. Bought audiences show cliffs, a flat line, a vertical jump of tens of thousands, then flat again, because followers were acquired in batches rather than earned over time.
If you can access the creator's follower history, those step-changes are diagnostic, especially when a jump in followers is not matched by any jump in engagement. Sudden growth with flat interaction is the signature of purchased followers, and it is visible in the shape of the data before you ever inspect an individual account.
Composition: who the followers actually are
Beyond rates and curves, the audience itself is inspectable. A creator's follower and following lists are public, and the proportion exhibiting bot characteristics, no posts, no photo, default-pattern usernames, extreme follow ratios, is the most direct measure of authenticity available.
Composition also speaks to fit and brand safety. An audience concentrated in a different geography or demographic than the creator claims, or an orbit of accounts you would not want your brand associated with, are findings the headline metrics hide. This is the layer that separates a real audit from an engagement-rate spot-check.
When the manual check isn't enough
For a small gifting arrangement, the 60-second test may suffice. For real budget, you want a number rather than an impression, what share of this audience is actually real, expressed as a percentage you can defend internally.
That requires profiling the full follower and following lists rather than sampling, which is what an audit tool does on public data. Serum pulls and profiles the entire audience and surfaces the fake-versus-real picture along with the associations that bear on brand safety, so you can vet a creator on public information before any money changes hands, and they are never notified you looked.
Frequently asked questions
How do I check if an influencer has fake followers?
Start with engagement rate, divide typical likes and comments by follower count, then read comment quality and inspect the growth curve. Consistently low engagement for the account's size, generic or bot-like comments, and sudden unexplained follower jumps are the classic fraud signals, and they are all visible without the creator's cooperation.
What engagement rate is suspicious for an influencer?
For an account pitched as an influencer, an engagement rate well under 1 percent is a yellow flag and under 0.5 percent is a red one, especially against a large follower count, since the platform median is already low and real influence shows up as proportionally higher interaction. Compare to the creator's tier, not a flat number.
What do fake influencer comments look like?
Generic, repetitive, emoji-only, or off-topic, 'Nice!', strings of emoji, comments unrelated to the post, frequently from accounts that themselves look fake. Authentic audiences leave specific comments referencing the actual content, so a hollow comment section under a large following is a strong tell.
Can I check an influencer without their permission?
Yes, for public accounts. Follower lists, following lists, and engagement are public, and U.S. courts have treated accessing no-login public data as lawful.[4] You are reviewing what anyone can see, and the creator is not notified.
How does the FTC factor into influencer deals?
The FTC's Endorsement Guides require influencers to clearly disclose material connections to a brand, and the agency has acted against both fake reviews and sellers of fake influence.[2][3] Vetting authenticity protects you from paying for fraud and from the compliance exposure of partnering with a creator who misrepresents their audience or fails to disclose.
Is a high follower count ever enough on its own?
No. Follower count is the cheapest metric to fake and the one fraud optimizes for. It tells you nothing about whether the audience is real or engaged, which is why diligence ignores the headline number and goes straight to engagement, comment quality, and audience composition.
Sources
- CHEQ & University of Baltimore (Roberto Cavazos), The Economic Cost of Bad Actors on the Internet: Influencers (2019)
- CBS News, coverage of the CHEQ / University of Baltimore influencer-fraud study
- U.S. Federal Trade Commission, Endorsements, Influencers, and Reviews (16 CFR Part 255)
- hiQ Labs v. LinkedIn (9th Cir.), case overview, Columbia Global Freedom of Expression
See what their profile won't tell you.
Serum pulls every account someone follows on Instagram or TikTok and profiles each one. No password. They are never notified.