How to spot bought followers and fake engagement

The checks, each with its limit

• Engagement-rate sanity. Pick ten ordinary, non-viral posts and compare likes plus comments to the follower count. Rough rules of thumb, which shift and decline with size: a small account often runs a few per cent, while a hundred-thousand-follower account pulling a fraction of a per cent looks inflated. Limit: dry or business niches and ad-driven feeds run lower naturally, and the benchmarks are ephemeral.
• Comment quality. Emoji spam, identical "nice pic" across posts, empty-shell usernames. Limit: real people leave short comments, and big accounts attract unpaid bot spam.
• Growth history. A free growth-history tool can reveal sudden unnatural spikes after flat periods, or a sawtooth of buy-then-purge. Limit: genuine virality spikes too, so cross-check against the posting history and any news.
• Audience-quality audits. Tools exist that estimate the share of fake or low-quality followers. Use them before a partnership. Limit: these are probabilistic model outputs, not proof, and they can misread niche, non-English or very local audiences as low quality.

The limit, and the decay tell

Check the same account more than once: a steady fake percentage reads as reliable, a sudden spike in followers with flat engagement is a purchase event. Two honest caveats hold the whole page together. The tools change, so re-confirm any you use and never treat a score as a verdict. And the deepest signal is structural: a bought audience does not engage and gets purged, which is why sellers offer "refill guarantees," a quiet admission that the platforms are deleting the fakes.