How do you check whether an influencer has fake followers?
In short
Compare engagement against follower count. Three to five percent is normal; an account with a million followers and a hundred likes on a post sits at 0.008% and is effectively dead. A suspiciously high figure is also a signal. Then check consistency across recent posts, the substance of the comments, and the follower growth history.
The number you start from
Engagement is the sum of interactions — likes, comments, shares and saves — divided by follower count. Calculate it across a dozen or more recent posts, never a single one.
| Result | What it means |
|---|---|
| Below 1% | Warning sign — bought audience or a burnt-out account |
| 1–3% | Low, but can be normal on large accounts |
| 3–5% | Normal for a healthy account |
| Above 10% | Also suspicious — check where the interactions come from |
The extreme case that circulates in the industry: a million followers and 80–100 likes per post. That is 0.008% against a 3–5% norm — a difference of several hundred times. Accounts like that don't need tooling to spot.
Red flags
- A step change in follower count over a short period with no post that could explain it.
- Repetitive, empty comments — emoji, "nice", "great" — rather than reactions to the content.
- Follower geography that doesn't match the language of the posts.
- A large gap between story reach and follower count.
- Likes that land in the same order of magnitude on every post regardless of subject.
What the numbers don't show
Read the last hundred comments. A real audience asks about details, argues, and refers back to earlier posts. Bought engagement is even and content-free. The exercise takes fifteen minutes and catches more than a tool report.
Check previous collaborations too. If the creator has promoted similar products and the posts carry comments like "ordered it, recommend", their recommendation demonstrably works. If there are many ad posts and none of them draw a reaction, the account is burnt out whatever the statistics say.
Tools
HypeAuditor and Social Blade are commonly used for profile audits — they show follower growth history and geographic distribution. Treat them as a first filter rather than a verdict: a tool cannot distinguish a dead account from one whose audience simply doesn't tap the heart.
What to do before you pay
Ask for screenshots of the last 30 days of analytics — reach, impressions, audience data. A creator who works seriously has them to hand. Refusal or delay is information in itself.
Even after the best vetting, the risk remains that the material never appears, or appears differently from what you agreed. That is why on Blogger Bank the budget goes into the Vault and is locked until you accept: the creator can see the funds reserved, and payout happens only once you have the material.
Related questions
- Does low engagement always mean bots?
- No. It can also mean a change of topic, or an audience that watches without reacting. Which is why you look at the trend, not one post.
- Does a verified profile mean no fake followers?
- Verification confirms identity, not audience quality. Two different things.
- How many posts should I review?
- Twenty recent ones is enough to read the trend; a hundred if the budget is large.
- What if the creator has bots but still sells?
- It happens when part of the audience is real. Settle it with a small test budget and a trackable discount code, not with an argument about statistics.
Sources
Vetting is half the protection
The other half is a payment that waits for your approval. Post a brief and pick a creator.
Create a free account