How do you check if an influencer has fake followers?

Start with engagement rate against the benchmark for that follower count — it is the fastest signal and it catches most cases on its own, because bought followers inflate the denominator without adding interactions. Then look at the shape of the engagement: generic comments, a like-to-comment ratio far outside the norm, follower spikes with no matching content event, and a follower list heavy on accounts with no photo and no posts. Any one of these can have an innocent explanation; three together rarely do.

Can you tell if followers are fake for free?

Yes, for the parts that matter most. Engagement rate against the size benchmark needs nothing but public numbers, and the comment section is public too. Our free authenticity check scores a YouTube or Twitch account on audience, engagement, growth and content signals with no login — and publishes how the score is calculated, which the paid tools generally do not.

What does a fake follower audit actually look at?

Four things. Audience authenticity — the share of followers with no profile photo, no posts, or bot-pattern usernames. Engagement quality — the rate against peers of the same size, and the like-to-comment ratio. Growth pattern — whether the follower curve has step-changes that no piece of content explains. And content signals — posting cadence and how much of the feed is sponsored. A tool that returns one number without showing you these is asking you to trust it.

How do you assess an influencer’s credibility before a brand deal?

Audience authenticity first, because everything else is priced off reach and an inflated audience makes every other number optimistic. Then engagement quality, then whether the audience matches yours in language and geography, then their sponsored-content history — a feed that is mostly ads converts worse regardless of how real the followers are. A low authenticity score is a data signal to investigate, not proof of fraud; treat it as a reason to ask questions before you sign.

How to Check if an Influencer Has Fake Followers

If you want to know how to check if an influencer has fake followers, the short answer is: you don’t need an expensive agency tool, you need to read the numbers the way a fraud analyst does. Bought followers leave a statistical fingerprint, and once you know where to look, a 10-minute audit surfaces the signals worth a closer look before a dollar leaves your budget.

Here’s the uncomfortable backdrop. HypeAuditor, citing the State of Influencer Marketing 2021, says over 55% of Instagram influencers were involved in fraudulent activities in 2020 — by HypeAuditor’s own definition of fraud. At the same time, some fake followers are normal: Modash notes that even legitimate creators pick them up because bots follow accounts on their own, and puts bigger creators at roughly 20–30% fake followers and creators under 50K at around 10–20%. So you’re not screening for zero bots — you’re screening for inflation big enough to change what you’re paying for.

The fastest signal: engagement rate vs. follower count

Real audiences engage. Bought ones don’t. The fastest first check is comparing an account’s engagement rate against published benchmarks for its size.

Those benchmarks vary a lot by source, so treat any one table as a reference point, not a pass mark. Here is HypeAuditor’s:

Follower tier (HypeAuditor’s bands)	Typical engagement rate
1K–5K (nano)	4%–5%
5K–50K (micro)	~2.5%–3.5%
50K–500K (mid-tier)	~2%–2.5%
500K+ (macro to mega)	~1%–2%

Source: HypeAuditor’s Instagram engagement rate guide (2025, updated 2026), based on HypeAuditor data. Other sources disagree: Modash’s help center (2026) gives 5–10% for 1K–10K, 3–7% for 10K–100K, 1–3% for 100K–1M and 0.5–2% for 1M+, while HypeAuditor’s own engagement calculator lists averages of 4.8% for 1,000–5,000 followers and 1.2% for 20,000–100,000. Niche and format move the numbers too, so compare a creator against accounts of similar size in the same niche.

There is no published “suspicious below” line, and we don’t invent one. A rate far under the range for its size is a reason to run the other checks below — not evidence on its own. No single metric proves an account bought followers.

A 220K account pulling 600 likes a post is at 0.27% (600 ÷ 220,000) — far below every range above, and a clear reason to keep checking. Run the math yourself with our free Instagram engagement rate calculator — paste in followers and average likes/comments and you’ll see instantly where they land.

7 checks worth running

Engagement rate is far below tier benchmark. Covered above. It’s a prompt to dig, not a finding.
Like-to-comment ratio is broken. There’s no reliable universal ratio, so compare against creators of similar size in the same niche. HypeAuditor notes that a significant gap versus similar accounts “might mean” likes or comments were inflated. 8,000 likes with 4 comments is worth asking about.
Follower spikes with no cause. Ask for a screenshot of follower growth (Instagram Insights shows it). A vertical jump of +20K in two days with no viral post, feature, or press behind it needs an explanation.
Comment quality is generic. ”🔥🔥🔥”, “Nice pic”, “Great content” repeated by accounts with no profile photo and egg-like usernames. Open five commenters — if three are empty shells, the engagement deserves scrutiny.
Follower geography doesn’t match the audience. A US fitness coach with 60% of followers in countries that don’t match their content or language should be asked why. Request audience-country data from their Insights.
Follower-to-following looks farmed. Accounts that follow 7,000 and have 7,200 followers may have grown through follow/unfollow tactics rather than genuine reach.
Story views are low even for the account’s size. Stories are harder to pad with a follower pack, but they reach a small share of followers on genuine accounts too — Hootsuite puts average Story reach at 2% of followers for brands with large followings (2026). Compare against similar creators, not against the raw follower count.

Why “follower count” is the wrong number entirely

Brands keep paying on follower count when what they actually want is genuinely engaged reach. Worked arithmetic: a creator with 40K followers at 5% engagement gets ~2,000 engagements per post (40,000 × 5%); a 400K account at 0.4% gets ~1,600 (400,000 × 0.4%). Price both on follower count and the second costs ten times as much for fewer engaged people.

This is the whole con of inflated accounts: they sell you a vanity number that doesn’t convert.

Run the audit in order

Start with the engagement-rate check because it’s fast and catches the most. If that passes, move to the like-to-comment ratio and a quick scroll through commenters. Only request Insights screenshots (growth curve, audience geography, story views) for creators who clear the first two and are getting real money. Asking a legitimate creator for their Insights is normal — refusal to share is itself a data point.

For a one-click sanity pass on a profile before you go deep, drop the handle into our free authenticity checker. It surfaces the same ratio and growth signals automatically so you know which accounts deserve the manual review.

Tools and manual checks, compared

You can audit by hand or with tools. Each has a place:

Method	What it catches	Limitation
Engagement-rate math	Bought followers (low rate)	Misses accounts that also bought engagement
Manual commenter scan	Bot comments, empty-shell accounts	Slow at scale; sample only 5–10
Insights screenshots	Growth spikes, geo mismatch, real reach	Requires creator cooperation
Automated checker	Ratio + growth anomalies fast	Best as a first pass, not the only check

The right workflow layers them: automated first pass to triage the shortlist, engagement math on survivors, then manual commenter scan and Insights request only on the finalists you’re about to pay. Don’t run the full manual audit on every name — it doesn’t scale, and the cheap checks narrow the list early.

A worked example

A brand is considering a 180,000-follower “fitness” account. Quick audit:

Average likes over last 10 posts: 720. Engagement rate: 720 ÷ 180,000 = 0.4% — far below HypeAuditor’s ~2%–2.5% range for 50K–500K accounts. Flag one.
Comments per post: ~6, mostly ”🔥” and “amazing.” Like-to-comment ratio and comment quality both poor. Flag two.
Story views (requested screenshot): ~1,100 — about 0.6% of followers (1,100 ÷ 180,000), low even next to Hootsuite’s 2% Story-reach average for large brand accounts. Flag three.

Three independent signals point the same direction. That still isn’t proof the creator bought followers — but it’s enough to ask for an explanation, and ~720 likes a post doesn’t justify a price based on 180,000 followers. Pass, or renegotiate to a rate that matches the audience you can actually see.

Reading the signals together

No single check is conclusive, and treating one as conclusive is how real creators get wrongly accused. What matters is which signals co-occur.

Pattern	Bought followers	Real but drifted audience	Real but under-distributed
Engagement rate vs. size band	far below	below	below
Comments per 100 likes	near zero	reduced but present	normal
Comment content	generic, off-topic, emoji-only	on-topic, from the wrong people	on-topic
Follower curve	step changes with no content event	flat or slowly declining	steady
Reach per post vs. followers	low	low	low
Follower geography vs. content language	mismatched	matches	matches

The three columns look identical on the one metric most people check — engagement rate — and diverge on every other row. That is the whole argument for running more than one check.

The practical rule: a low rate alone is a question. A low rate plus near-zero comments plus an unexplained follower step is a pattern that justifies passing or asking hard questions. Two of the three, and you ask the creator directly — a genuine one will have a story that matches the curve, like a post that broke out or a feature that drove a spike.

And state it carefully. “This account’s engagement is well below the benchmark for its size” is a defensible observation. “This creator bought followers” is an accusation about a real person, and unless you can see their purchase records you cannot support it. A low score is a signal to investigate, not a verdict.

When the numbers are borderline

Not every low engagement rate means fraud. A few legitimate reasons an account underperforms:

Reels-heavy accounts show lower like rates because views, not likes, are the currency. Weight your read toward saves, shares, and comments.
Huge accounts naturally decay. Engagement rate falls as follower count rises; Modash’s 1M+ range is 0.5%–2%, so a 2M-follower account at 0.8% sits inside it.
Niche B2B audiences lurk more than they tap.

The point isn’t to reject everyone below a threshold — it’s to make the creator explain the gap with real data. A genuine creator can usually do that quickly from their own Insights.

Vet before you pay, not after. A 10-minute audit on a five-creator shortlist is cheap next to paying for an audience that isn’t there.