Audit any X influencer before you pay
Bought views come with matching likes, so follower counts and engagement rate won't catch them. Run any X / Twitter handle, creator or KOL: real reach, who actually replies, engagement pods, fake followers, and what a post costs per 1,000 views.
Runs on live X data. Add their price to see if it’s fair; leave it empty to get a fair range.
Three tells follower counts hide
Bought views come with matching likes, so the engagement rate looks fine. These are what gave away every inflated account we calibrated the audit on.
Likes nobody talks under
Real audiences leave 11 to 22 replies per 100 likes. Bought likes and views arrive in bulk and silent: rings we've checked sit at 4 to 7.
The same repliers, every post
In an engagement pod a few dozen accounts reply under each other's posts. On organic accounts the regulars write 0–16% of replies; in the rings we checked, 46–80%.
Ads without an #ad
Few paid posts carry a label. They show up as a brand the feed keeps naming, a campaign hashtag, a tracking link, or a “try it here” link in the author's first reply. One feed we checked was 46% ads with almost no #ad.
The audit reads all three — check a handle ↑
How the score and fair price work
Seven checks add up to a score out of 100. Two red flags (an engagement pod or bought likes) cap it at 40, whatever the rest says. The fair price runs on the numbers that survive.
Median views of recent original posts against followers, judged by size: a 1M account seeing 2–3% is healthy, a 50k account seeing under 1% has an audience that isn't there.
Organic accounts we measured get 11–22 replies per 100 likes. Rings that buy views and likes sit at 4–7. Bought engagement is cheap to fake as likes and expensive to fake as conversation.
We read the replies under five recent posts. Friends replying to each other is normal: on organic accounts the regulars write 0–16% of replies. When they write 45%+, the circle is the conversation (an engagement pod), and the report names them. Silence fails too: 100+ likes and nobody replying is as bad a sign as a pod.
Likes, replies and reposts per view, against the ~1% X norm. On its own it misses bought views, which is why it carries the least weight.
The weakest original posts of an organic account still get 25–55% of a typical post's views; replies in the author's own threads are left out. A swing to under 10% means some posts were pushed and the rest died.
A sample of the latest followers, checked for empty accounts (no posts, default avatar, days old). Only the share above what's normal for the account's size counts.
Sponsor wording, tracking links, campaign hashtags and brands the feed keeps promoting, even without an #ad. A feed that's half ads lowers both the score and the fair price.
The fair price
We fit the list prices of 147 X creators in our network against their numbers. Follower count explains price best, then niche, then typical views: for the same audience and reach, AI creators charge the most, trading accounts about the same, crypto accounts a little over half. The audit plugs in the followers and views it believes, after removing empty accounts and discounting pods, bought likes and boosted posts, then takes 20–35% off when a third to half of the feed is already ads. Add the creator's quote and it tells you whether it's a good deal, market price or overpriced, and what to offer.
Unlike a classic fake follower checker, it doesn't stop at the follower count. Most inflated accounts we see today buy views and likes, not followers, and those only show up in how few people reply and in who those people are.
Influencer audit FAQ
How do I check if an X influencer has fake followers or fake engagement?
Run their handle in the audit above. It reads their recent posts, the replies under them and a sample of their newest followers, then scores real reach per post, engagement rate, replies against likes, whether the same accounts reply every time (an engagement pod), whether views are boosted on some posts only, how many new followers are empty accounts, and how much of the feed is already paid promotion.
Why isn't engagement rate enough to spot a fake influencer?
Because the people selling views sell likes with them. An account that buys 30,000 views per post usually buys the 200 likes that make the engagement rate look normal. What they rarely buy is conversation: replies from different, real people. That's why the checker weighs replies against likes and looks at who is replying.
What is an engagement pod on X?
A group of accounts that agree to like and reply to each other's posts so the algorithm pushes them further. The numbers are real clicks, but they come from the same circle every time, so a brand paying for a post reaches the pod, not an audience.
How much should I pay an X influencer per post?
Price what's real: audience size, niche and views, after the fakes are taken out. We fit the list prices of 147 X creators in our network: follower count weighs most, then niche (AI creators charge the most, crypto about half as much for the same numbers), then typical views. The checker runs that on the followers and views it believes, so pods, bought likes and boosted posts lower the fair price. Add their quote and it tells you whether it's a good deal, market price or overpriced, and what to offer.
What is a Twitter audit?
A Twitter (X) audit checks whether an account's numbers are real: how many followers are empty or bought, whether posts actually get seen, and whether the likes and replies come from an audience or from a pod. Brands run one before paying an influencer; creators run one on themselves before pitching. This tool is a Twitter audit built for the brand side: it ends with a verdict and a fair price.
How accurate is the audit?
We calibrated it on accounts we know: organic crypto and tech accounts scored 83–100, and a network of creators that trade engagement scored 40–44, caught by replies vs likes and by who replies, rather than by engagement rate. It reads public data, so a quiet week or a viral post can move a score; treat it as a screen before a test post, not a final judgment.
Does the account I check get notified?
No. The checker only reads public posts, replies and followers. Nothing is followed, liked or messaged, and no login is involved.
Is the influencer audit free?
Yes. No login, no limits, any public X account. It reads public data only, so the account you check is never notified.
Hire creators who already pass
NotPeople runs managed X and Reddit campaigns with vetted creators. Inorganic activity is cut from the report and from performance payouts.