💬 TikTok Comments Scraper

Pull every public comment from any TikTok video into clean, structured data — JSON, CSV, or Excel — in seconds. ⚡

✨ What you get

🎯 Every public top-level comment — handle, nickname, comment text, likes, replies, timestamps, language, pinned status, author profile URL. 🚀 Multiple videos in one run — feed it a list of URLs, get one consolidated dataset back. 🧹 Clean, flat records — ready to drop into a sheet, a database, or a notebook. No HTML, no nested noise. 🪙 Pay-per-result pricing — you only pay for the comments you actually receive. Predictable, capped via maxItems. 🔁 Incremental refresh mode — re-run with since to pick up only the new stuff. No re-scraping the whole comment list. 📦 Three export formats — JSON, CSV, and Excel, all generated automatically.

🚀 Quick start

1. Click "Try for free" on the Apify Store page.
2. Paste your TikTok video URLs into the Video URLs field — one per line.
3. Press Start. Watch the live log; comments stream into the dataset as they're collected.
4. Download the dataset in JSON, CSV, or Excel from the run page.

That's it. No accounts, no API keys, no signing up to anything else.

📥 Input

Field	Type	Default	Description
videoUrls	string[]	— required	List of TikTok video URLs or numeric video IDs. Accepts https://www.tiktok.com/@user/video/<id>, plain <id>, or any URL containing /video/<id>.
maxCommentsPerVideo	integer	100	Stop scraping each video after this many top-level comments. Set 0 for no per-video cap.
maxItems	integer	0	Hard cap on total comments across all videos. 0 disables the global cap. Use to control pay-per-result spend.
since	datetime	—	ISO timestamp like 2026-04-01T00:00:00Z. Skips comments older than this. Use for incremental re-runs.
concurrency	integer	3	How many videos to scrape in parallel. Range 1–8. Lower if hit by rate limits, higher on premium proxies.

📤 Sample output

{
  "url": "https://www.tiktok.com/@nike/video/7631536776661765389",
  "id": "7631763288914133776",
  "videoId": "7631536776661765389",
  "videoUrl": "https://www.tiktok.com/@nike/video/7631536776661765389",
  "text": "Supporting that ballroom wow I love Brooks best running shoes ever!",
  "createdAt": "2026-04-21T18:39:42Z",
  "likeCount": 45,
  "replyCount": 2,
  "isPinned": false,
  "isAuthorPinned": false,
  "isLikedByAuthor": true,
  "language": "en",
  "parentCommentId": null,
  "images": [],
  "user": {
    "id": "7324198047750980613",
    "uniqueId": "jennicwo46s",
    "nickname": "JenniC",
    "secUid": "MS4wLjABAAAAO2o-FYv8q-XwUL0hQRYrbwjCm04E3mg5Tq8VQJ1SNkxJdgWefQ1dJRLg6oMH-v5D",
    "avatarUrl": "https://p16-common-sign.tiktokcdn.com/...avatar.jpg",
    "verifiedReason": null,
    "profileUrl": "https://www.tiktok.com/@jennicwo46s"
  },
  "scrapedAt": "2026-05-10T12:34:56Z"
}

💼 Use cases

Audience	What they do with it
🛍️ Brand & social teams	Track sentiment on competitor posts. Catch complaints early. Find what audiences love (or hate) about your campaigns.
🤝 Influencer agencies	Vet creators by reading actual comment quality — not just like counts. Spot bot-engagement patterns.
📈 Market research	Mine emerging trends, slang, product mentions, and audience demographics from comment streams.
🎯 Performance marketers	Find high-intent buyers who actively engage with niche-relevant content; build look-alike audiences.
🎓 Academic researchers	Discourse analysis, language datasets, network science on social engagement patterns.
🤖 ML/NLP teams	Build labeled training data for sentiment, topic, hate-speech, or recommendation models.

💡 Tips & tricks

• Start with maxItems: 100 to see what the data looks like before doing a big run. The first record alone tells you whether the fields you need are there.
• Use since for daily refreshes. Set it to the timestamp of your last run; the scraper skips everything older. Cuts re-scraping costs to zero on stable videos.
• For very viral videos (10k+ comments), consider running them one-at-a-time with concurrency: 1 — you'll get cleaner runs and avoid rate-limit retries.
• Batch your URL list. A single run with 100 URLs is more cost-efficient than 100 runs with 1 URL each.

❓ FAQ

Q: Are reply threads included? A: This run captures top-level comments only. Each record carries replyCount so you know how many replies exist. Reply-thread collection is on the roadmap (see Changelog).

Q: What about comments from private or restricted accounts? A: The scraper only sees what's publicly accessible on the TikTok web frontend. Private accounts, age-restricted videos, and geo-restricted videos are out of scope.

Q: Can I scrape comments without a TikTok login? A: Yes. This actor doesn't require any TikTok login or token from you.

Q: What if a video has 50,000 comments? A: Use maxCommentsPerVideo to cap it. Scraping the full tail of a viral video takes time and money — sample what you need.

Q: Why is the avatar URL not loading after a day? A: TikTok signs CDN URLs with an x-expires query param. Avatars typically stop loading ~24h after the run. Re-scrape, or save your own copy at run time if you need them long-term.

Q: Can I resume a failed run? A: Apify retains the dataset across run attempts. Re-run with since set to the last scrapedAt you saw, and de-dupe on id in your downstream pipeline.

⚖️ Legal

This actor scrapes only publicly accessible content from TikTok video pages — the same data that any unauthenticated visitor can read. You are responsible for using the resulting data in compliance with TikTok's Terms of Service, the GDPR, the CCPA, and any other laws or platform rules that apply to you. Comments are user-generated content and may contain personal data; treat them as PII when storing or processing. If you publish or republish scraped data, attribute it to TikTok and to the original commenters per the platform's content rules.

📜 Changelog

v0.1 — initial release

• Top-level comment extraction across one or many TikTok videos
• Per-video and global item caps
• Incremental mode via since filter
• Per-comment record with author profile, engagement, language, timestamps
• JSON, CSV, and Excel exports
• Built-in deduplication by comment id