TikTok Influencer Scraper

An Apify actor that scrapes TikTok creators by hashtag, extracts profile data + emails, and outputs a clean list ready for influencer outreach.

---

What It Does

1. Hashtag → Creators — Navigates to https://www.tiktok.com/tag/{hashtag}, intercepts TikTok's internal API calls, and collects creator handles while scrolling to load more content.
2. Profile Data — Visits each creator's profile (@handle) and intercepts the user-detail API response to capture followers, following, likes, bio, and bio link.
3. Email Extraction — Regex-scans the bio text for email addresses, and optionally follows bio links (Linktree, Beacons, solo.to, etc.) to find hidden emails.
4. Filter + Normalize — Applies a min_followers filter and outputs a clean 14-field schema.

No fragile TikTokApi library used. This actor uses Playwright with stealth settings to behave like a real browser.

---

Supported Use Cases

• Influencer outreach — Find creators in a niche, get their emails directly
• Market research — Understand which creators dominate a hashtag
• Agency prospecting — Build creator lists by vertical or geography
• Competitor analysis — See who's posting in your space

---

Input Schema

Field	Type	Default	Description
hashtags	string[]	required	Hashtags to scrape (without #). E.g. ["dutchfashion", "sportmode"]
min_followers	integer	5000	Skip creators with fewer followers
max_results	integer	1000	Maximum total creators returned
extract_emails	boolean	true	Regex-scan bio text for email addresses
follow_bio_links	boolean	true	Follow Linktree / Beacons / solo.to links to find hidden emails
country_hint	string	""	Optional ISO country code to annotate results (e.g. "NL", "DE") — best-effort, not a hard filter

Example Input (input.json)

{
  "hashtags": ["dutchfashion", "sportmode"],
  "min_followers": 5000,
  "max_results": 1000,
  "extract_emails": true,
  "follow_bio_links": true,
  "country_hint": "NL"
}

---

Output Schema (14 fields)

Field	Type	Description
handle	string	TikTok username (without @)
display_name	string	Full display name
first_name	string	First word of display_name, emojis stripped
followers	integer	Follower count
following	integer	Following count
likes	integer	Total likes (heart count)
bio	string	Bio text
email	string	First email found in bio or bio link page
bio_link	string	URL in the creator's bio
profile_url	string	https://www.tiktok.com/@{handle}
platform	string	Always "tiktok"
country_hint	string	Passed through from input
hashtag_source	string	Which hashtag surfaced this creator
scraped_at	string	ISO 8601 UTC timestamp

Example Output Record

{
  "handle": "fashiongirl_nl",
  "display_name": "Sophie 🌷 Fashion",
  "first_name": "Sophie",
  "followers": 48200,
  "following": 312,
  "likes": 1450000,
  "bio": "Dutch fashion & lifestyle 🌷 collabs: sophie@fashiongirl.nl",
  "email": "sophie@fashiongirl.nl",
  "bio_link": "https://linktr.ee/fashiongirl_nl",
  "profile_url": "https://www.tiktok.com/@fashiongirl_nl",
  "platform": "tiktok",
  "country_hint": "NL",
  "hashtag_source": "dutchfashion",
  "scraped_at": "2026-03-03T14:31:00Z"
}

---

Pricing

Component	Rate
Actor start	$0.00005
Per result	$0.003
1,000 creators	≈ $3.00

---

How It Works (Technical)

Scraping Strategy

Rather than using the fragile TikTokApi library (which breaks frequently), this actor uses Playwright with Chromium to:

1. Navigate to https://www.tiktok.com/tag/{hashtag}
2. Intercept XHR/fetch responses matching TikTok's internal API patterns:
   • /api/challenge/item_list/
   • /api/post/item_list/
   • /api/search/item/full/
3. Scroll the page to trigger lazy-loading and collect more creator handles
4. For each unique creator, navigate to https://www.tiktok.com/@{handle} and intercept /api/user/detail/
5. Run email extraction on bio + optionally follow bio_link

Anti-Detection

• Realistic Chrome user-agent string
• navigator.webdriver property hidden
• playwright-stealth patches applied
• Random delays between requests (0.5–2s)
• No automation flags in browser args

Email Extraction

email_extractor.py handles two strategies:

1. Bio regex — standard [a-z]+@[domain] pattern, with spam filtering
2. Bio link follow — httpx (not Playwright) fetches the linked page and scans its HTML

Supported bio link aggregators: Linktree, linktr.ee, Beacons, solo.to, Bio.link, Campsite, lnk.bio, AllMyLinks, WithKoji, Carrd, About.me, TapLink.

---

Local Development

Prerequisites

pip install playwright playwright-stealth httpx apify apify-client beautifulsoup4
playwright install chromium

Run Locally

1. Create input.json in the project root:

{
  "hashtags": ["dutchfashion"],
  "min_followers": 1000,
  "max_results": 20,
  "extract_emails": true,
  "follow_bio_links": true,
  "country_hint": "NL"
}

2. Run the actor:

$python src/main.py

3. Results are written to output.json in the project root.

Docker (matches Apify environment)

docker build -t tiktok-scraper .
docker run --rm \
  -v $(pwd)/input.json:/usr/src/app/input.json \
  -v $(pwd)/output:/usr/src/app/output \
  tiktok-scraper

---

Notes & Limitations

• TikTok changes its internal API frequently. The XHR interception patterns may need updating if TikTok changes its API endpoint structure.
• Rate limiting — TikTok may throttle or CAPTCHA at high request rates. Add delays or use Apify proxies for large runs.
• Email coverage — Only creators who publicly share their email (in bio or bio link pages) will have emails populated. Expect ~10–30% email coverage depending on the niche.
• Country hint — TikTok does not expose user location in most API responses. country_hint is passed through from your input, not scraped from TikTok.

---

License

MIT