Douyin Scraper (抖音): Videos, Profiles, Search & Comments

What does Douyin Scraper do?

Douyin Scraper extracts data from Douyin (抖音), the Chinese version of TikTok: keyword search results, creator profiles and their latest videos, single videos, and comments. You get likes, comments, shares, collects (saves), hashtags, publish time, duration, author details, music, and no-watermark video links, ready for a spreadsheet, database or AI pipeline.

No Douyin account, no cookies and no Chinese phone number needed. Run it in Apify Console, call it from the API, schedule it, or connect it to Make, Zapier, n8n or Google Sheets.

What data can you extract from Douyin?

🎬 Video ID and URL	📝 Description	#️⃣ Hashtags
❤️ Likes	💬 Comment count	🔁 Shares
⭐ Collects (saves)	⏱️ Duration	📅 Publish time
📥 No-watermark MP4 link	🖼️ Cover image	🎵 Music / original sound
👤 Author name and Douyin ID	🔗 Author profile URL	🖼️ Image-post (图文) photos
👥 Followers and following	🏆 Total likes	🎞️ Video count
✅ Verified badge	📍 IP location (IP属地)	🪪 Bio (signature)
🗨️ Comment text	👍 Comment likes	↩️ Reply count

Why use this Douyin scraper?

• Everything in one actor: search, profiles, videos and comments, so you don't need four separate tools.
• Works from outside China. Douyin blocks most foreign and proxy traffic; this actor uses a dedicated data service that gets through reliably.
• No-watermark MP4 links for every video, plus cover images and the original sound.
• Creator profile data: followers, following, total likes, video count, bio, verification and IP location (IP属地).
• Comments with IP location and like counts, for sentiment and audience research.
• Free filters: only recent videos, or only videos above a like threshold, skipped before anything is charged.

Common uses: China market research, KOL / influencer discovery for brands entering China, trend and hashtag tracking, e-commerce and product research, competitor monitoring, and training data for AI models.

How to scrape Douyin

1. Click Try for free and open the Input tab.
2. Add search keywords (Chinese or English), creator profile URLs, video URLs, or any mix.
3. Set how many videos you want and, optionally, how many comments per video.
4. Click Start, then download the results as JSON, CSV, Excel or HTML.

Input

{
    "searchKeywords": ["咖啡", "美妆教程"],
    "profiles": ["https://www.douyin.com/user/MS4wLjABAAAAk9mpFofCPQpsv1I1SVFxdb-A_BWiJnGy1qoDQxLEB4s"],
    "videoUrls": ["https://www.douyin.com/video/7534679152504376595"],
    "maxResultsPerSearch": 20,
    "maxVideosPerProfile": 20,
    "commentsPerVideo": 0,
    "sortBy": "relevance",
    "publishedWithin": "any",
    "postedWithinDays": 0,
    "minLikes": 0
}

Accepted links: douyin.com/video/<id>, links with ?modal_id=<id>, v.douyin.com share links (copy them from the Douyin app's Share button), iesdouyin.com share pages, and profile links douyin.com/user/<sec_uid>.

Output

Every row has a rowType: video, comment, profile or error. The dataset has separate Videos, Comments and Profiles views.

Video row (shortened):

{
    "rowType": "video",
    "id": "7534679152504376595",
    "url": "https://www.douyin.com/video/7534679152504376595",
    "description": "雨天开车秒变“睁眼瞎”？玻璃起雾别慌！3步极速除雾法",
    "hashtags": ["雨天开车除雾正确方法", "新手司机"],
    "createTime": "2025-08-04T10:43:20.000Z",
    "durationSec": 67.8,
    "likes": 12761,
    "comments": 649,
    "shares": 5135,
    "collects": 6914,
    "authorName": "懂车小彬",
    "authorUrl": "https://www.douyin.com/user/MS4wLjABAAAA...",
    "videoUrl": "https://v26-colds.douyinvod.com/...",
    "coverUrl": "https://p3-pc-sign.douyinpic.com/...",
    "musicTitle": "@懂车小彬创作的原声",
    "source": "url"
}

Comment rows have text, likes, replies, ipLocation, createTime and the commenter. Profile rows have nickname, douyinId, followers, following, totalLikes, videoCount, signature, ipLocation and verified.

How much does it cost to scrape Douyin?

Pay per result, no subscription. See the Pricing tab for exact prices:

• one charge per video,
• a smaller charge per comment,
• one charge per profile info row,
• a small start fee per run.

Errors and filtered-out videos are free. Higher Apify plans get up to 20% off.

Run it on a schedule

Save a task with the creators or keywords you follow, set Only videos from the last N days to 1 or 7, and add a daily schedule. Connect Google Sheets, Slack or a webhook to get new Douyin videos automatically.

Use it with AI agents (MCP)

Add the Apify MCP server (https://mcp.apify.com) to Claude, ChatGPT or Cursor and ask, for example: "Find the 20 most-liked Douyin videos about 咖啡 this week and summarize the trends."

FAQ

Is there an official Douyin API? Douyin's Open Platform only gives access to accounts that authorize your app, and requires a Chinese business entity. This actor reads public data about any video, creator or search without that.

What are the alternatives? Other Douyin scrapers on Apify (usually one per data type), Chinese data vendors, or scraping manually from a mainland IP. This actor combines search, profiles, videos and comments in one run.

Why are view counts missing? Douyin hides play counts on most videos. Likes, comments, shares and collects are always included.

Can I download the videos? Each video row has videoUrl, a direct no-watermark MP4 link. Links expire after a few hours, so download them soon after the run.

Does it work for 图文 (image posts)? Yes: isPhotoPost is true and imageUrls lists the images.

Is it legal? The actor only reads public content. Respect copyright, Douyin's terms and privacy laws such as PIPL and the GDPR.

Related actors

• Douyin Transcript Scraper: turn Douyin videos into text (Mandarin speech-to-text).
• TikTok Profile Scraper: the same data for TikTok creators.
• TikTok Keyword Video Search: search TikTok videos by keyword.
• Video Transcript Scraper: transcripts from TikTok, Instagram, YouTube and X.
• Reddit Scraper: posts and comments from Reddit.

Privacy

This Actor sends anonymous run inputs (such as keywords and URLs) and a hashed account ID to the developer to see which inputs are useful. Tokens and secrets are never sent. Nothing from your results is collected.