Engagement Rate & Fake Follower Checker: Instagram, TikTok & YouTube

You scraped 500 creators. Which ones are worth paying? Feed this Actor the rows your Instagram, TikTok or YouTube scraper already gave you and get back, for every creator: engagement rate, how it compares with accounts the same size, warning signs of bought followers, a 0 to 100 score, a grade and a plain shortlist / check / skip verdict. Best creators first.

No scraping, no login, nothing to break. It reads the data you already have, so it is fast and cheap: $0.003 per creator scored.

Who it is for

• Brands and agencies building influencer shortlists for a launch, Black Friday or a seasonal campaign.
• Influencer platforms and lead-gen builders who scrape creators in bulk and need to rank them automatically.
• AI agents that find creators and need a quick, explainable "is this one real?" answer per account.

What goes in

Any mix of these, all at once. Posts and profiles of the same creator are merged on platform + username.

Source	Works with
Instagram profile scraper rows	apify/instagram-profile-scraper (followers, following and the latest 12 posts per profile)
Instagram post rows	apify/instagram-scraper in posts mode (add the profile run too for follower counts)
TikTok video rows	clockworks/tiktok-scraper and look-alikes (authorMeta with fans, plus plays, likes, comments, shares)
YouTube video rows	streamers/youtube-scraper and look-alikes (subscribers, views, likes)
Your own sheet	A CSV, Excel file or Google Sheet with username, platform, followers and either posts or avgLikes / avgComments / engagementRate

Pass Datasets (pick one or more scraper runs), a File or Google Sheet URL, or Creator rows as JSON.

What comes out (one row per creator)

Field	Meaning
engagementRate	(likes + comments, + shares on TikTok) per post, as a % of followers
engagementRateByViews	(likes + comments + shares) per video, as a % of views. TikTok and YouTube are judged on this
medianEngagementRate	Same, using the median post, so one viral post cannot flatter the number
tier, benchmarkEngagementRate, engagementLevel	Size band (nano, micro, mid, macro, mega), the typical rate for that band and platform, and very low / low / typical / high / unusually high
riskSignals	Warning signs, each with a plain sentence: low engagement for the size, likes without comments, views without likes, low views, follow-for-follow, uneven likes, unusually high engagement, inactive, mostly sponsored, few posts
authenticityScore	100 minus points for each warning sign
creatorScore, grade	0 to 100 from engagement vs typical (45%), authenticity (35%) and posting rhythm (20%); A to F
verdict, verdictReason	shortlist, check or skip, with the reason
avgLikes, avgComments, avgViews, avgShares, avgSaves, medianLikes	The averages behind it
postsPerWeek, daysSinceLastPost, sponsoredPostShare, topHashtags	Rhythm and content
nicheMatch, nicheWordsFound	Whether your niche words appear in the bio, captions or hashtags
confidence	high (8+ posts and a follower count), medium, low

Example row (shortened):

{
  "handle": "nasa", "platform": "tiktok", "followers": 2200000, "postsAnalyzed": 7,
  "avgLikes": 13584, "avgComments": 377, "avgViews": 228043,
  "engagementRate": 0.66, "engagementRateByViews": 6.35, "tier": "mega",
  "benchmarkEngagementRate": 5, "engagementLevel": "typical",
  "riskSignals": [], "authenticityScore": 100, "creatorScore": 84, "grade": "A",
  "verdict": "shortlist", "verdictReason": "Shortlist: typical engagement for a mega account and no serious warning signs."
}

The creator's own columns from your scraper (profile fields, not every post) are kept next to the new ones. Turn Keep original columns off for a compact table. A run summary with verdict, grade and warning-sign counts is saved as OUTPUT in the key-value store, and you can also save a CSV or Excel file.

Filters

Set Minimum / Maximum followers, Minimum engagement rate or Niche words (fitness, vegan, skincare...). Creators that do not pass are left out and not charged. Rows to keep can write only the shortlist.

Pricing

• $0.003 per creator scored (a creator that comes back with an engagement rate and a verdict).
• $0.01 per CSV or Excel file, only if you ask for one.
• Not charged: private accounts, creators with no follower count, rows with no username, and anyone your filters leave out.

Scoring 1,000 creators costs about $3. You also pay your scraper as normal; this Actor adds no scraping cost of its own.

Use it after your scraper (no code)

On your scraper's Integrations tab, add this Actor and keep Datasets as ["{{resource.defaultDatasetId}}"]. Every scraper run is scored the moment it finishes.

For AI agents: the Actor is on Apify's MCP server (mcp.apify.com/?tools=nerolabs/engagement-rate-checker) and can be paid per call with x402.

How honest is it?

• Every number comes from your rows. Nothing is guessed or looked up elsewhere.
• Benchmarks are typical industry ranges per platform and size: a rough guide, not a law.
• A warning sign is a reason to look closer, not proof that followers are fake. Some real creators have quiet comment sections or one viral post. The reason for every sign is written on the row so you can judge it.
• Some YouTube scrapers write 0 comments when they did not collect them; when every video shows 0, comments are treated as unknown rather than as a warning.

FAQ

My Instagram post rows say "needs_followers". The posts scraper does not include follower counts. Add the Instagram profile scraper's run for the same accounts to Datasets and the two are merged.

Which engagement rate formula is used? Instagram: (likes + comments) / followers. TikTok: (likes + comments + shares) / views, with the per-follower rate also given. YouTube: (likes + comments) / views. The formula used is written in engagementBasis on every row.

Pinned posts? Left out when there are at least 4 other posts, because they are often old hits.

Can it check a creator's followers one by one? No. It judges the audience from engagement patterns, which is faster and works on any scraper's output. For a follower-by-follower audit you need a tool that scrapes each follower.

If this saved you from paying the wrong creator, a review on the Apify Store helps a lot.