TikTok Influencer Authenticity Scorer

Analyze TikTok influencers at scale to detect fake followers, bot comments, and engagement pods. Get authenticity scores and partnership risk recommendations powered by AI comment analysis.

🚀 Quick Start Guide (New Users)

Get your first report in under 2 minutes. No coding required.

1. Try Demo Mode (No API key needed): Go to the Input tab, enter a TikTok username (e.g., charlidamelio), ensure Demo Mode is checked, and click Start. Check the Output tab to see a sample report.
2. Run Real Analysis (BYOK): Uncheck Demo Mode. Get a free API key from platform.openai.com or console.anthropic.com, paste it into the AI API Key field, enter your target usernames, and click Start.
3. Export Your Data: Once the run finishes, go to the Output tab and click Export to download your reports as CSV, JSON, or Excel.

What does this Actor do?

This Actor analyzes public TikTok influencer profiles and returns an authenticity score (0–100) with detailed suspicious signal reports. It scrapes recent posts and comments, applies AI-powered sentiment and bot detection analysis, and produces a clear partnership risk recommendation.

It does not access private data, DMs, or non-public information. It only analyzes what the influencer has chosen to share publicly.

Why use this Actor?

Influencer fraud costs brands an estimated $1.3 billion annually. 63% of brands say fake followers are their top influencer marketing concern. Manually vetting influencers is slow, subjective, and does not scale.

This Actor automates:

• Fake follower detection — analyzes engagement patterns and follower velocity
• Bot comment detection — uses AI to score comment authenticity
• Engagement pod detection — checks commenter diversity across posts
• Partnership risk scoring — gives you a clear recommendation (Low/Medium/High risk)

What data can it extract?

Field	Type	Description
username	string	TikTok username (without @)
followers	integer	Total follower count
authenticityScore	integer	0–100 score (higher = more authentic)
engagementRate	string	(likes + comments + shares) / followers
genericCommentRatio	number	% of comments that are generic/emoji-only
commenterDiversityIndex	number	Higher = more diverse commenters (better)
suspiciousSignals	array	List of detected red flags
recommendation	string	Low/Medium/High risk partnership recommendation
analyzedAt	string	ISO timestamp of analysis

How much does it cost?

This Actor uses pay-per-event pricing. You pay per influencer analyzed, not per compute unit. Free tier users can run up to 10 influencers per month at no cost.

For AI comment scoring, you bring your own API key (BYOK). AI costs are typically $0.01–$0.03 per influencer analyzed, depending on the number of comments processed.

Input Parameters

• usernames (required): Array of TikTok usernames to analyze
• maxPostsPerUser: Number of recent posts to scrape (default: 10)
• maxCommentsPerPost: Comments to analyze per post (default: 50)
• aiApiKey: Your OpenAI or Anthropic API key (required when demoMode is false)
• aiProvider: Select openai or anthropic
• demoMode: Run with sample data without an AI key (default: true)

📊 Output Sample

You can download your dataset in JSON, CSV, Excel, or HTML formats. Here is exactly what the output looks like:

JSON Format:

{
  "username": "@example_creator",
  "followers": 250000,
  "authenticityScore": 34,
  "engagementRate": "1.2%",
  "genericCommentRatio": 0.62,
  "commenterDiversityIndex": 0.45,
  "suspiciousSignals": [
    "Follower spike of 15,000 on March 3, 2026",
    "62% of comments are generic/emoji-only",
    "Same 18 accounts comment on every post"
  ],
  "recommendation": "HIGH RISK — Do not partner without further due diligence",
  "analyzedAt": "2026-10-15T14:30:00Z"
}

CSV Format (Spreadsheet view):

username	followers	authenticityScore	engagementRate	genericCommentRatio	recommendation
@example_creator	250000	34	1.2%	0.62	HIGH RISK — Do not partner

Tips & Advanced Options

• Start with Demo Mode to test the Actor without an AI key.
• Increase maxPostsPerUser to 20+ for more accurate engagement analysis.
• Use the Apify API to run this Actor programmatically from your own applications.
• Connect via MCP — this Actor is available as an MCP tool for AI agents:

{
  "mcpServers": {
    "apify": {
      "type": "http",
      "url": "https://mcp.apify.com/?tools=fetch-actor-details,YOUR_USERNAME/tiktok-influencer-authenticity-analyzer"
    }
  }
}

FAQ

How does the Actor detect fake followers? It analyzes three signals: engagement rate relative to follower count, follower velocity anomalies, and comment quality/diversity. When multiple signals overlap, the risk score increases.

Do I need an AI API key? Only if you turn off Demo Mode. Demo Mode uses cached sample data. For real analysis, you need an OpenAI or Anthropic API key.

Is my API key secure? Yes. The aiApiKey input field uses isSecret: true, which means it is encrypted and never logged or displayed.

Can I analyze private TikTok accounts? No. This Actor only analyzes public profiles.

How many influencers can I analyze per run? There is no hard limit. For 100+ influencers, consider splitting into multiple runs to manage compute costs.

What if an influencer has very few posts? The Actor requires at least 3 recent posts for meaningful analysis. If an influencer has fewer, the report will flag insufficient data.

Disclaimers and Support

Our Actors are ethical and do not extract any private user data, such as email addresses, gender, or location. They only extract what the user has chosen to share publicly. You should not scrape personal data unless you have a legitimate reason to do so.

For issues, use the Issues tab in Apify Console. For programmatic access, see the API tab.