TikTok Super Scraper

One TikTok scraper for every public surface: accounts, videos, keyword search, comments, related videos, the Explore feed, hashtag reach, the TikTok Ad Library and TikTok Shop, all in a single run and a single dataset.

This TikTok scraper replaces the usual stack of five or six separate tools plus the manual work of joining their outputs. Give it any mix of handles, keywords, video links, advertisers or shop terms and it returns one merged table, every row tagged with the source it came from, so you can pull a creator's videos, the comments under them, the ads their competitor is running and matching Shop products in the same job.

What you get

• Account data: handle, display name, follower/following/likes counts and video count for any public TikTok account.
• Videos: caption, play/like/comment/share/save counts, duration, sound, hashtags and posting date, per account, per keyword, per related seed, or from the public Explore feed.
• Comments: one row per comment (and reply, when enabled) with text, like count, reply count and author.
• Related videos: for any seed video, the videos TikTok surfaces as related, about sixteen per seed, each a full record, to widen coverage from a starting point.
• Hashtag reach: how many videos carry a hashtag and how many views they have between them.
• Ads: every ad an advertiser runs in the TikTok Ad Library, with format, first/last shown, impressions and spend where TikTok publishes them.
• Shop products: title, sale and original price, currency, brand, rating, reviews, units sold, images and seller for TikTok Shop listings.

Every row carries a source and source_actor field, so a mixed run stays one groupable table rather than several you have to reconcile by hand.

Why scrape TikTok

TikTok has no public API for most of this. The official research and business APIs are gated behind approval, cover a fraction of the fields, and rate-limit hard. Everything a logged-out visitor can see on the web, an account's videos, a video's stats, the comments, a hashtag's reach, the Ad Library, Shop listings, is public but scattered across different pages and different internal endpoints, each returning a different shape. Pulling a competitor's content strategy, a creator's engagement history, or an advertiser's live campaigns normally means running several tools and writing glue code to line them up on the same key.

This actor does that lining-up for you. One input, one dataset, one billing event per row you receive. You manage no proxies, solve no challenges, and write no per-endpoint parsers.

Two ways to run it

Auto mode. Set query to a single term and it fans out to every term-driven source at once, the same string is used as the account handle, the search keyword, the advertiser name, the shop keyword and the hashtag. One input gives you the account, its search landscape, the ads that brand runs and matching Shop products in one dataset.

Flexible mode. Leave query empty and fill only the per-source fields you want. Each source runs on its own input and only the fields you fill are run.

The two combine: with a query set, any per-source field you also fill overrides the query for that source (e.g. query: "nike" plus advertisers: ["Nike Inc"] searches videos/shop/profile for "nike" but the Ad Library for "Nike Inc"). The id-based sources (comments, downloads, related) always need explicit video links or ids, a free-text query cannot drive them.

Input

Provide a query (auto mode) and/or any combination of the fields below (flexible mode). Only the sources you give input for are run.

Field	Type	What it does
query	string	Auto mode: one term used as handle, keyword, advertiser, shop term and hashtag at once
profiles	array	Handles or profile links; returns each account and its recent videos
keywords	array	Keyword search; returns matching videos per keyword
comment_videos	array	Video links/ids; returns comments (and replies) per video
download_videos	array	Video links/ids; returns the media file/URL and quality ladder
advertisers	array	Advertiser names; returns ads from the TikTok Ad Library
shop_keywords	array	Keywords; returns TikTok Shop products
hashtags	array	Hashtags; returns reach figures
related_to	array	Seed video links/ids; returns related videos per seed
explore	integer	How many videos to pull from the public Explore feed (0 disables)
explore_category	string	Optional Explore category id
sources	array	Limit the run to named sources: profile, search, comments, downloads, ads, shop
region	string	Two-letter country code for the market (e.g. US, NL)
max_per_source	integer	Cap on rows fetched per source call
include_replies	boolean	When reading comments, also fetch replies
concurrency	integer	How many source calls to run at once

Output

The default dataset holds one tiktok_super_summary row (rows per source, which sources returned data, and any per-source errors) followed by one row per result, each tagged with its source. Example video row:

{
  "source": "profile",
  "source_actor": "tiktok-profile",
  "video_id": "7106594312292453675",
  "url": "https://www.tiktok.com/@nasa/video/7106594312292453675",
  "description": "Launch day.",
  "created_at": "2024-05-31T12:00:00Z",
  "play_count": 581000,
  "like_count": 98900,
  "comment_count": 1289,
  "share_count": 358,
  "handle": "nasa",
  "follower_count": 95900000,
  "hashtags": ["space"]
}

The summary and the error list are also written to the run's key-value store.

Use cases

Competitor content intelligence. Marketing and social teams give the actor a competitor's handle plus a set of keywords and get the competitor's recent videos, the engagement on each, and the videos ranking for those keywords in one table. Grouping by source separates the account's own output from the wider keyword landscape, so you can see both what they post and what they compete against.

Ad and campaign monitoring. Growth and paid-media teams pass a list of advertisers and get every ad each runs in the TikTok Ad Library, with formats and, where TikTok publishes them, impressions and spend. Run it on a schedule to track when a competitor launches or retires a campaign.

Creator research and outreach. Agencies and brand teams pass a set of handles and get follower counts, video output and per-video engagement, the numbers you need to size a creator before a partnership, without opening TikTok forty times.

Product and price tracking on TikTok Shop. E-commerce teams pass shop keywords and get products with prices, ratings, review counts and units sold, for catalogue research or competitor price monitoring in the market you set with region.

How it compares

The Apify Store has strong single-purpose TikTok scrapers, most focused on one surface: a profile scraper, a comment scraper, a Shop scraper. This actor's difference is breadth in one run: profiles, search, comments, related, explore, hashtags, ads and Shop behind a single input and merged into one dataset. If you only ever need one surface, a single-purpose scraper is a fine choice. If you routinely pull several and join them yourself, this removes that work and the per-tool billing that comes with it.

Pricing

Pay per event. You are charged only per result row delivered, across every source, at $0.0003 per result. There is no run fee, and the run summary row is not billed. Free-plan runs return up to ten rows per run. You only pay for results you receive, with no per-compute-unit charges.

Limits and gotchas

• Explore is best-effort. TikTok's Explore feed sometimes requires more than a plain request; when it declines, the run reports the reason on that source rather than returning an empty feed. Every other source is dependable.
• Hashtags return reach, not the video list. TikTok does not serve a hashtag's videos to logged-out visitors, so the hashtag source returns counts and says so, rather than an empty list.
• Login-gated data is out of scope. A logged-out visitor's view is what you get; private accounts, follower lists and anything behind a login are not available.
• Set region to the market you want. Prices, currency and some feeds vary by country; pin it rather than relying on a default exit.
• max_per_source is the cost ceiling. It caps rows per source call; raise it for depth, lower it to keep runs cheap.
• Cold start. The first call in a run adds a few seconds while the sources spin up.

FAQ

Can I scrape TikTok without an API key? Yes. You provide only the handles, keywords or links you want; no TikTok credentials are needed.

Can I get TikTok videos and their comments in one run? Yes. Put the account or video links in profiles/comment_videos and both come back in the same dataset, tagged by source.

Does it scrape the TikTok Ad Library? Yes, via the advertisers field, one row per ad an advertiser runs.

Can I scrape TikTok Shop prices? Yes, via shop_keywords, with price, rating, reviews and units sold per product.

Can I limit which sources run? Yes. Fill only the fields you need, or set sources to an explicit list.

Related Actors

• TikTok Profile Scraper
• TikTok Ad Library Scraper
• TikTok Comment Scraper
• TikTok Shop Scraper