X (Twitter) Fake Follower Auditor - Audience Quality Check

Audit any X (Twitter) account for fake followers and bots. The actor samples the account's most-recent followers, scores every sampled follower on proven bot signals, and returns a fake-follower percentage, an audience quality grade, and the full per-follower evidence - the same audit influencer-marketing platforms charge hundreds per month for.

What you get

One audit summary:

• Fake follower % and suspicious % from a real follower sample (not an estimate)
• Audience quality grade (A-D)
• Signal breakdown - how many sampled followers had a default profile picture, zero tweets, suspicious follow ratios, spammy handles, empty bios, or brand-new accounts
• Sample statistics - blue-verified count, no-tweets %, default-avatar %, average follower/following counts across the sample

Plus one record per sampled follower:

• Bot score (0-100) with a likely_fake / suspicious / likely_real verdict
• The exact signals that fired, follower/following/tweet counts, account creation date, and profile URL

Export everything to JSON, CSV, Excel, or Google Sheets from the Apify console.

Use cases

• Influencer vetting - check an influencer's audience authenticity before signing a sponsorship deal
• Detecting bought followers - a sudden wave of new, empty accounts in the recent-follower sample is the classic signature of purchased followers
• Agency audits - screen creator rosters and report audience quality to clients
• Competitor audience analysis - compare audience quality across accounts in your niche
• Brand safety - verify partner accounts aren't inflated with bot audiences

How it works

The actor samples up to 5,000 of the target account's most-recent followers and scores each one on bot signals, each with a plain-language meaning:

• Default profile picture - the account never uploaded an avatar, the strongest low-effort-bot tell
• Zero tweets - the account has never posted; bots that only exist to follow rarely tweet
• Suspicious follow ratio - following thousands of accounts while having almost no followers is classic follow-bot behavior
• Brand-new account - created within the last 30 days (strong signal) or 90 days (weak signal); bot farms churn out fresh accounts
• Spammy handle - machine-generated username patterns like long trailing digit runs (user48291047)
• Empty bio and no cover picture - no effort spent making the profile look human
• Flagged automated - the account is publicly labeled as automated on X

Each fired signal adds its weight to the bot score (capped at 100). A score of 50+ is likely_fake, 25-49 is suspicious, below 25 is likely_real. The fake-follower percentage is the share of sampled followers scoring as likely fake.

How to use

1. Enter the X username to audit (with or without @)
2. Choose a Follower sample size (200 is a good default; 1000+ for large accounts)
3. Run the actor - the summary is the first record in the Dataset tab, follower evidence follows

Output format

Summary record:

{
  "recordType": "summary",
  "auditedUsername": "example",
  "auditedUserId": "123456789",
  "followerCount": 152340,
  "sampleSize": 200,
  "fakeFollowerPct": 18.5,
  "suspiciousPct": 12.0,
  "qualityGrade": "B",
  "signalBreakdown": { "default_profile_picture": 31, "zero_tweets": 44, "suspicious_follow_ratio": 12, "empty_bio": 58 },
  "sampleStats": { "blueVerifiedCount": 3, "noTweetsPct": 22.0, "defaultAvatarPct": 15.5, "avgFollowers": 412, "avgFollowing": 890 },
  "sampledNewestFirst": true,
  "checkedAt": "2026-07-17T02:00:00.000Z"
}

Follower record:

{
  "recordType": "follower",
  "auditedUsername": "example",
  "followerUsername": "user48291047",
  "followerUserId": "987654321",
  "fullName": "User 48291047",
  "botScore": 65,
  "verdict": "likely_fake",
  "flags": ["default_profile_picture", "zero_tweets", "spammy_handle", "no_cover_picture"],
  "followers": 3,
  "following": 1840,
  "tweets": 0,
  "isBlueVerified": false,
  "hasDefaultAvatar": true,
  "accountCreated": "Mon Jun 22 09:14:03 +0000 2026",
  "profileUrl": "https://x.com/user48291047"
}

Free tier limits

Runs on a free Apify plan are limited to 25 followers sampled per run (the sample size input is capped automatically). Upgrade to any paid Apify plan to remove the limit and audit with full sample sizes.

FAQ

Do I need to log in or provide cookies? No. No login, no cookies, no risk to your own account.

How accurate is it? The sample is drawn from the account's live follower list, newest first. At 200 followers the fake percentage is typically within a few points of a full audit; increase the sample size for tighter confidence.

Why newest first? Bought followers arrive in bursts, so the most-recent followers are exactly where purchased or bot audiences show up first.