TikTok Scraper

Thirdwatch portfolio: 8,800+ users across 140+ public Actors, 2M+ records delivered, >95% run success. Explore all Thirdwatch Actors.

Find TikTok videos and user profiles by keyword or handle — descriptions, public engagement counts, creator info, music, and tags.

What you get

Structured video and user data without a login, collected over plain HTTP (no browser) so runs default to 256 MB and finish in seconds. @username queries list the creator's latest videos from TikTok's own embed page (plus a public index for deeper history) and read every video's public page for exact view, like, comment, share and bookmark counts. #hashtag queries start from TikTok's hashtag embed listing; keyword and creator searches use a public index for discovery, then hydrate each hit from TikTok's page state. Fallback rows are labeled explicitly instead of presenting unavailable engagement as real zeroes.

Output fields (videos)

Field	Description
id	TikTok video ID
description	Video caption
url	Direct link to the video
authorName	Creator display name
authorUsername	Creator handle (without @)
authorUrl	Profile URL
authorFollowers	Creator follower count
authorVerified	Verified-creator flag
viewCount	Number of views
likeCount	Number of likes
commentCount	Number of comments
shareCount	Number of shares
bookmarkCount	Number of saves
duration	Video duration in seconds
createdAt	ISO 8601 upload timestamp
musicTitle	Audio/music title
musicAuthor	Audio/music artist
thumbnailUrl	Thumbnail image URL
hashtags	Hashtags extracted from the caption
isAd	Paid-promotion flag
dataSource	Which public index produced this row — useful when auditing coverage gaps
dataQuality	full_video, partial_indexed_engagement, partial_tiktok_page, or discovery_only
engagementAvailable	true only when public engagement metrics were actually available

Output fields (users)

Field	Description
id	TikTok user ID
username	Handle
nickname	Display name
url	Profile URL
bio	Profile biography
verified	Verified flag
followers	Follower count
following	Following count
likes	Total likes received
videoCount	Number of posted videos
avatarUrl	Profile picture URL

Example output

{
    "id": "7345678901234567890",
    "description": "Easy 15-minute pasta recipe #cooking #recipe #foodtok",
    "url": "https://www.tiktok.com/@chefmike/video/7345678901234567890",
    "authorName": "Chef Mike",
    "authorUsername": "chefmike",
    "authorFollowers": 1250000,
    "authorVerified": true,
    "viewCount": 8500000,
    "likeCount": 425000,
    "commentCount": 3200,
    "shareCount": 18000,
    "bookmarkCount": 95000,
    "duration": 47,
    "createdAt": "2026-03-15T14:30:00Z",
    "musicTitle": "original sound",
    "musicAuthor": "Chef Mike",
    "hashtags": ["cooking", "recipe", "foodtok"],
    "isAd": false
}

Input parameters

Parameter	Required	Description
queries	Yes	List of search terms, @usernames, or #hashtags. Examples: ["cooking", "@gordonramsay", "#recipe"].
searchType	No	videos or users. Default videos.
maxResults	No	Max total results across all queries. Default 8.
maxResultsPerQuery	No	Max results per individual query. Default 50.
liveEnrichment	No	Opens a hardened browser session only for rows the fast request path could not complete. Default false; select at least 2048 MB memory when enabled (the default 256 MB is enough for every HTTP-only run).
sortBy	No	For @username video queries: default keeps TikTok's own order (pinned posts may come first); newest picks the most recent uploads before the result limit is applied. Hashtag and keyword queries are unaffected. Default default.
downloadVideos	No	Save each video's MP4 (up to 25 MiB) to the run's key-value store. Adds download status and file details to matching rows; a failed download never drops the video's metadata. Video searches only. Default false.
maxAgeDays	No	Only keep videos with a known publication date within this many days. 0 disables the filter. Default 0.

Use cases

• Influencer research: Follower counts, engagement rates, and posting cadence for any creator.
• Competitor tracking: Growth and top-performing content for a roster of competitor accounts.
• Trend monitoring: Pull hashtag and keyword feeds to surface viral content early.
• Brand monitoring: Find videos mentioning your brand, product, or category.
• Partnership discovery: Discover creators in a niche for brand deals.
• Market research: Analyze audience behavior across topics and formats.

Use cases & recipes

Step-by-step guides on thirdwatch.dev/blog:

• Monitor Brand TikTok Presence (2026 Guide)
• Research TikTok Creator Engagement (2026 Guide)
• Scrape TikTok Profiles and Videos at Scale (2026)
• Track TikTok Trending Content by Hashtag (2026)

-end

Limitations

• Private accounts cannot be scraped.
• Video media URLs are short-lived and may expire shortly after extraction.
• Very new or low-engagement videos may have incomplete metadata.
• A creator's newest ~10 videos come straight from TikTok; anything beyond that is discovered through a public index, which can lag TikTok and returns older videos in index order rather than newest-first.
• When exact-page hydration is blocked, the row is retained as dataQuality: "discovery_only" (index hit) or "partial_embed" (embed listing, rounded view count only) with engagementAvailable: false; unavailable numeric metrics are not evidence of zero engagement.
• Rate limits apply on heavy runs — spread large jobs across queries.

Compared to alternatives

• vs. clockworks/tiktok-scraper ($0.0037/result, 156K users): That actor is cheaper with a broader surface area. This actor keeps the schema tight and consistent across video and user search and is maintained on a predictable release cadence.

Pairs well with Instagram Scraper and YouTube Scraper for cross-platform creator research.

FAQ

Can I scrape a specific creator's videos? Yes. Prefix the handle with @ in queries (e.g., @gordonramsay). Switch searchType to videos for their video feed or users for the profile record.

Does hashtag search work? Yes. Prefix the term with # (e.g., #recipe) and use searchType: "videos".

Are view counts accurate in real time? Counts are taken from the public source selected for the run. Use engagementAvailable and dataQuality to distinguish complete live rows from index-only discovery rows.

Do I need an API key or login? No.

Last verified: 2026-07

More scrapers at thirdwatch.dev.