Paste any Instagram handle to estimate how much of the audience is real, dormant, or bot-heavy. Free, no card.
Estimates how much of an audience is real, dormant, or bot-heavy. It reads public engagement patterns — it cannot prove that anyone bought followers.
Create a free account to see the full breakdown.
We store your name and email so you can come back to your audit history.
TheTechHacker Creator Audit · built by Rahul Vithala · thetechhacker.com
What the audit measures, what it can prove, and what it can't.
It estimates how much of an Instagram or YouTube audience is real, dormant, or bot-heavy. You give it a handle, and it reads the account's public engagement patterns — likes, comments, views, follower growth, comment text — and scores them against what is normal for an account of that size. You get six scores and a plain-language explanation of anything that looks off.
No, and it will never say that it can. Fake-audience detection is inference, not proof. The only party who can prove anything is the platform itself, because only the platform sees the account data. What this tool does is spot patterns that bought audiences reliably produce — and then tell you how strong that pattern is, so you can decide what to do. That is why every output says estimated risk.
Five checks, in order of how much they tell you. One: divide average likes plus comments by follower count — if a 100k account gets 400 likes, something is wrong. Two: read the comments. Real audiences argue, ask questions, reference the actual post; fake ones say "nice pic" and "🔥". Three: open twenty followers at random and look for empty profiles — no photo, no posts, following 3,000 people, followed by nine. Four: look for follower jumps that no post explains. Five: check whether the audience is in the market the creator claims to reach. This tool automates all five and weighs them together, which matters, because no single one of them is conclusive on its own.
It depends heavily on size — smaller accounts engage harder. As rough bands on Instagram: under 10k, roughly 2.5–9%; 10k–50k, roughly 2–6.5%; 50k–200k, roughly 1.6–5%; 200k–1M, roughly 1.2–4%; above 1M, roughly 0.9–3%. On YouTube the equivalent question is what share of subscribers watch a median video — above 8% is healthy, and below 1% is a serious flag. The tool compares each account against the band for its own tier rather than one universal number.
This is the distinction most tools get wrong, and it is the one that matters most commercially. A fake audience was purchased: empty profiles, generic comments, growth spikes no post explains. An inactive audience is real people who followed years ago and stopped showing up. Both produce low engagement, so a single-metric tool flags them identically — but one is fraud and the other is just an old account. The audit scores them separately, so a dormant creator gets a high inactive score and a low bot score.
Authenticity is the headline, 0–100, weighing everything. Bot risk rises only with signals that indicate manipulation — empty follower profiles, fake comments, unexplained spikes. Inactive audience rises when engagement is thin but the manipulation signals are clean. Engagement quality, growth risk and comment authenticity break out the three signals people most often want to see on their own.
It is a weighted blend, not a single metric. On Instagram: engagement quality 25%, growth pattern 20%, follower profile quality 20%, comment quality 15%, like-to-comment ratio 10%, geography consistency 10%. On YouTube: views-to-subscriber ratio 25%, growth pattern 20%, comment quality 15%, engagement quality 15%, view anomalies 15%, upload consistency 10%. Multi-signal scoring is used because any single metric can be explained away — several drifting at once is much harder to explain.
90–100: highly authentic. 75–89: mostly authentic, with some inactive followers — normal for most established accounts. 60–74: suspicious, worth a manual review before you spend money. Below 60: high estimated risk; do not commit budget without account-level insights from the creator directly.
A jump far above the account's own baseline that no content explains. Real spikes have a cause you can see — a video that went wide, a collaboration, press. Purchased followers arrive in a block on an ordinary week with nothing behind it. The tool stores a daily follower snapshot for every account it audits, so the longer an account has been tracked, the sharper this signal gets. On a brand-new audit it says so, rather than pretending to know.
Four things, measured together: generic text that would fit under any post ("nice", "great post", "amazing"); duplicates, where the same phrasing repeats across posts; emoji-only replies at high volume; and repeat commenters — the same handful of accounts on every single post, which is the signature of an engagement pod. Any one of these is normal in small amounts. All of them at once is not.
No profile photo, no posts, no bio, a username with a long random number stuck on the end, and a wildly lopsided follow ratio — following two thousand accounts while being followed by eleven. The tool samples followers and counts how many show three or more of these traits at once. One trait means nothing; plenty of real people have no bio. Three at once is a different story.
The audit itself runs before you sign up — you see the account, how many posts were scanned, and how many flags were raised. The account is what unlocks the scores and the findings. It also lets you come back to your audit history, and it keeps automated scripts from burning through the API quota that makes the tool work.
Instagram has no public API for arbitrary handles, and scraping it violates the platform terms and breaks constantly. The legitimate route — Graph API business discovery — reaches Instagram Business and Creator accounts only. If a handle is a personal profile, no compliant tool can audit it, and any tool that claims otherwise is scraping. YouTube is different: the official Data API covers every public channel.
Because the honest answer is that we cannot see them. Instagram's public API returns posts, likes and comments — it never returns a follower list or audience geography. Those two signals only become available if the creator shares their own insights export. When they are missing, the tool scores them neutral and labels them unavailable, rather than guessing and passing the guess off as a measurement.
Treat it as a triage tool, not a verdict. It is good at telling you which accounts deserve a closer look and which do not, and the more signals that agree, the more confident you can be. But a low score is an estimate built on public patterns, and there are innocent explanations for most individual signals. Use it to decide where to dig — then ask the creator for account-level insights before any final call. And do not publish a score as an accusation: "this account bought followers" is a claim about a real person, and it is not one this data can support.