TikTok Creator Search

🔎 TikTok influencer search by niche, keyword, or name

This TikTok influencer search tool helps you discover public TikTok creators by creator name, username, niche, brand, or keyword. Enter up to 10 search terms and collect matching creator profiles in a structured Apify dataset—without a TikTok login, browser, phone, or emulator.

Turn searches such as fitness coach, skincare, OpenAI, or a username into ranked creator profiles. Every row retains the query that found it for easier review and shortlisting.

Recent videos are included at no extra charge. Enable includeRecentPosts to attach each creator's latest available videos, then read them in the Recent content and Recent videos output views. Normal creator-result and optional profile-enrichment charges still apply.

✨ What can this TikTok influencer search tool do?

• 🔍 Search creators your way: Use names, usernames, niches, topics, brands, or keywords.
• 📚 Run multiple searches together: Submit 1–10 queries in a single run.
• 🎯 Choose your target size: Request 20–1,000 creators per query. TikTok may return fewer when fewer matching accounts are available.
• 👤 Enrich public profiles: Optionally add bios, bio links, public counters, region, and full profile images.
• 🎬 Collect recent posts at no extra charge: Optionally attach creators' latest videos, including captions, dates, public metrics and links.
• 📖 Review videos comfortably: Expand videos within a creator row or switch to a readable per-video table with dates, engagement metrics and watch links.
• 🧹 Keep results organized: Duplicate suggestions are removed within each query. Optional unique mode combines overlapping creators across queries and retains every matched query and rank.
• 📤 Export creator data: Download JSON, CSV, Excel, XML, or another Apify-supported format.
• ⚙️ Automate repeat searches: Run the Actor through Apify Console, the API, client libraries, schedules, or integrations.

🚀 How to search for influencers on TikTok

1. Add a clear niche or keyword to Search queries, such as vegan recipes, personal finance, or gaming creator.
2. Add more queries to explore several niches, brands, or known creators in one run.
3. Choose a target from 20 to 1,000 creators per query.
4. Turn on Enrich profiles when you need public bios, bio links, full counters, region, and larger profile images.
5. Turn on Include recent posts to review creators' actual recent publications. Choose the maximum number of videos per creator.
6. Click Start, review the ranked results, and export or connect the dataset to your next workflow.

Use specific phrases such as London fitness coach or sustainable fashion creator for more focused results. A query guides TikTok's suggestions; it is not a location or audience filter.

📥 Input

Field	Required	What it controls
searchQueries	Yes	1–10 creator names, usernames, niches, brands, or keywords.
maxCreatorsPerQuery	No	Target number of creators for each query, from 20 to 1,000. Default: 20.
outputMode	No	perQuery (default) keeps separate query matches; unique returns one row per creator across queries.
maxCreators	No	Run-wide unique-creator output limit, 1–10,000. Default: 1,000. Used only in unique mode.
enrichProfiles	No	Adds available public profile details. Default: false.
includeRecentPosts	No	Includes latest videos in recent_feed at no extra charge. Default: false.
maxRecentPostsPerCreator	No	Maximum latest videos per creator, 1–500. Default: 12. Availability and request limits may produce fewer.

Example search

{
  "searchQueries": ["fitness coach", "sustainable fashion"],
  "maxCreatorsPerQuery": 100,
  "enrichProfiles": true
}

Unique creators across queries

{
  "searchQueries": ["fitness coach", "personal trainer"],
  "outputMode": "unique",
  "maxCreatorsPerQuery": 100,
  "maxCreators": 150,
  "enrichProfiles": true
}

Unique mode completes discovery across the configured queries before enrichment and output. It selects the first-discovered creators up to maxCreators, merges matches from the executed searches, and enriches and writes each selected creator once. Discovery remains bounded by maxCreatorsPerQuery, the unique target and the remaining charge budget, so provenance covers only the results actually fetched. Fewer unique creators may be available than requested.

Each row retains query and search_rank for its first discovery and adds matchedQueries, queryRanks and queryCoverage. Open the Unique creators dataset view to inspect them. A creator matching multiple queries incurs one creator-result charge and at most one successful profile-detail charge. The default perQuery mode preserves separate rows and charges for creator-query matches; maxCreators has no effect in that mode.

Partial results, retries and resume

Discovery retries transient transport failures, HTTP 408/5xx and empty responses on the failing page, with bounded backoff. HTTP 401/403/429, rejected API responses and malformed schemas are reported without repeated retries. An exhausted page preserves its earlier creators, and a failed query does not prevent later queries from running. Each query has a 100-page safety cap.

OUTPUT.queryResults includes status (ok, partial, error, or running), stopReason, accepted pages, attempted requests, retries, and discoveredCount. apiHasMore retains the upstream flag even when a repeated page triggers no_new_creators. count attributes saved rows to their first query; matchedCreatorCount also counts unique rows matched by subsequent queries. OUTPUT.failedQueries lists queries that produced no usable results after a search error.

Accepted discovery pages are checkpointed in SEARCH_STATE. Platform persistence/migration events also save state. Restarting the same run/storage resumes unfinished discovery using saved creator IDs, then reconciles recordKey values against the existing dataset before enrichment and writes. Previously delivered rows are skipped. Completed partial/error queries retain their status; start a new run to retry those queries. Use a fresh dataset and key-value store for a new search: a checkpoint or dataset from another input/version is rejected.

Every mode checks the remaining charge budget before discovery and enrichment. Profile-enabled runs reserve the worst-case combined creator/profile price, which can stop conservatively when only a base row would fit. Dataset writes recheck the budget. OUTPUT is updated during the run and on exit, including interrupted runs. In unique mode, dataset rows begin after discovery.

Dataset delivery and the platform's profile-detail charge are separate operations. Recovery skips already delivered rows rather than replaying an uncertain charge. detailsCharged reports the SDK charging count; per-query charge counts are null when they cannot be reconciled with saved profiles. Checkpointed request metrics can omit work interrupted before the next checkpoint.

Search with recent videos

{
  "searchQueries": ["OpenAI"],
  "maxCreatorsPerQuery": 20,
  "includeRecentPosts": true,
  "maxRecentPostsPerCreator": 12,
  "enrichProfiles": true
}

Only includeRecentPosts and maxRecentPostsPerCreator configure recent content. The defaults are off and 12 videos (range 1–500), respectively. Every discovered creator is eligible: no shortlist, follower, keyword or date filter is applied. The collector requests the latest available videos and manages 10 workers, a page budget derived from the requested count and session pacing internally. No media files are downloaded; permanent video-page links are returned.

Profile enrichment and recent posts share one profile request when both are enabled. Disabling recent posts preserves the existing search/profile-enrichment behavior. Removed experimental input fields are ignored, including when supplied through the API.

After discovery, feed-enabled results stream in completion order; sort by query and search_rank for discovery order. queryResults[].recentFeedSeconds measures the feed-and-save stage in per-query mode, including dataset writes. Unique mode reports this combined stage in recentFeedSecondsTotal; per-query timings are null because the shared batch can span several originating queries. profileEnrichmentSecondsTotal measures separate profile-only requests; shared feed/profile work is included in the feed timing. There is no cross-run cache.

Recent-feed output and status

After a run, open Output and select the view that fits your task:

View	What you can read
Quick overview	Creator profiles, followers, account flags and original search rank.
Recent content	One row per creator, collection status, video count and an expandable list of videos.
Recent videos	One displayed row per video: creator, caption, date, watch link, views, likes, comments, shares, saves, duration and hashtags.

Switching views does not run another scrape or add charges. Recent posts must be enabled for new runs to collect videos; the views do not retroactively enrich earlier results.

Posts are nested in recent_feed.items inside the existing creator row. The Actor does not create separate post rows. Use Recent content for creator status/counts and an expandable videos field. Switch to Recent videos to read one video per displayed row, with its caption, publication date, views, likes, comments, shares, saves, duration, hashtags and clickable video links. Creator profile links remain available in Recent content. This view expands the nested data for reading only: it does not create or bill additional records. Both views are static schema tabs; the video table has content only when videos were collected. Creators with unavailable or empty feeds remain visible in Recent content. Export the original JSON for full nested records. Each post includes id, url, post_type, description, hashtags, created_at, create_time, author, author_match, stats, is_pinned, and available text-language/duration data. Missing metrics are null, not invented zeroes. Video files and temporary CDN media links are not collected.

Field	Meaning
recent_feed_status	ok, partial, or unavailable
recent_post_count	Number of returned videos after deduplication
recent_feed.stop_reason	Why collection stopped: requested_limit, api_reported_end, page_limit, or a specific failure
recent_feed.pagination	Requested/accepted pages and the API's has_more flag
recent_feed.collected_at	Time of the actual collection, also preserved for cached results
recent_feed.cache_hit	Whether this row reused a run-local result

ok means the requested limit or API-reported end was reached; it is not proof of a complete account history. A valid empty feed can contain zero posts. partial preserves available records when collection stops early; unavailable means no trustworthy selection was obtained. A feed failure never removes the creator's discovery row.

The collector verifies the profile against the original search candidate's stable ID/secUid and verifies the author and date of every post. It checks chronological order, handles pinned posts without jumping over recent items, and deduplicates pagination. It never substitutes keyword-search hits for a profile feed. Captions/hashtags are evidence for a later relevance check, not automatic proof that a creator matches the searched niche; no LLM scoring or transcription is performed here.

Concurrency, traffic and memory

• Workers reuse isolated HTTP sessions and sticky proxy connections. Sessions never cross worker threads.
• Submission is bounded to the configured worker count, not one queued future per discovered creator.
• Feed-enabled results stream in completion order; search_rank preserves discovery order.
• A run-local five-minute cache deduplicates identical creator identities across queries. It is capped at 128 entries and 16 MiB of serialized payload; Python object overhead adds memory beyond that cap. In per-query mode, rows retain their original query: cache reuse avoids HTTP calls, not dataset-row charges. Unique mode deduplicates before enrichment and billing.
• The remaining pay-per-event budget caps the number of eligible rows before feed work begins. Completed feed rows are saved immediately; saved post arrays are released from query aggregates. Hosted per-row write latency has not yet been benchmarked; the live performance figures come from local SDK storage in the actual Docker Actor.
• Each worker waits at least one second between requests. Pages stay sequential within a creator.
• A rate limit, verification request, login requirement or HTTP 403 pauses new feed requests for the run. Already in-flight requests may finish; other creators retain explicit unavailable/partial statuses. Transient errors are not retried by the feed collector, and private profiles are not accessed.

The OUTPUT summary includes post/status totals, combined collection time, actual feed HTTP requests, cache hits, peak worker concurrency and any pause reason. request_count inside a cached feed describes the original fetch; use OUTPUT.recentFeedMetrics.requests for total network requests in this run. When both features are enabled, profile request time is part of recentFeedSeconds, not measured again as a separate profile pass (profileEnrichmentSeconds is then null). Start with the default concurrency; more workers are not always faster.

📤 What TikTok profile data is returned?

The Actor saves one row per discovered creator. Some counters can be empty in TikTok's suggestion response; enrichment requests more public profile fields when available.

Data group	Available fields
Search context	Original query and result rank
Creator identity	TikTok user ID, secure user ID, username, nickname, profile URL, and avatar
Profile details	Bio, bio link, profile image, and region when enrichment succeeds
Audience and activity	Followers, following, total likes received, and video count when available
Account signals	Verified, business, private, TikTok Shop, seller, and live-status flags; null when unknown
Enrichment status	Whether profile enrichment was requested, enriched, partial, or unavailable

Explicit zero counters remain 0; missing or invalid counters remain null. Missing or unrecognized account flags remain null, while an explicit negative flag remains false. Profile-only enrichment verifies available stable user IDs/secUid against the discovery row before merging. Conflicting or unverified identities retain the discovery data and set profile_enrichment_status: "unavailable" with profile_enrichment_reason of profile_identity_mismatch or profile_identity_unverified; these rows incur no profile-detail charge.

Example search output

With enrichProfiles set to false, each discovery result has the following shape. TikTok may omit counters from its suggestion response, in which case their values are null.

{
  "query": "fitness coach",
  "id": "1234567890123456789",
  "sec_uid": "MS4wLjABAAAA-example-secure-user-id",
  "unique_id": "coachmaya",
  "nickname": "Maya | Fitness Coach",
  "url": "https://www.tiktok.com/@coachmaya",
  "avatar": "tos-useast5-avt-0068-tx/example-avatar.jpeg",
  "follower_count": null,
  "video_count": null,
  "verified": false,
  "business": true,
  "private": false,
  "shop": false,
  "live": false,
  "search_rank": 1,
  "profile_enrichment_status": "not_requested",
  "profile_source": null
}

Example enriched profile output

With enrichProfiles set to true, a successful profile lookup enriches the same discovery row. The values below are illustrative; fields that are not public or available may be null.

{
  "query": "fitness coach",
  "id": "1234567890123456789",
  "sec_uid": "MS4wLjABAAAA-example-secure-user-id",
  "unique_id": "coachmaya",
  "nickname": "Maya | Fitness Coach",
  "url": "https://www.tiktok.com/@coachmaya",
  "avatar": "https://p16-sign-va.tiktokcdn.com/example-avatar.jpeg",
  "avatar_uri": "tos-useast5-avt-0068-tx/example-avatar.jpeg",
  "signature": "Simple strength and mobility tips",
  "bioLink": "https://example.com/coachmaya",
  "follower_count": 125400,
  "following_count": 310,
  "total_likes_received": 2300000,
  "video_count": 412,
  "profile_image_url": "https://p16-sign-va.tiktokcdn.com/example-avatar.jpeg",
  "profile_region": "US",
  "profile_source": "profile-embedded-json",
  "verified": false,
  "business": true,
  "private": false,
  "shop": false,
  "live": false,
  "search_rank": 1,
  "profile_enrichment_status": "enriched"
}

💡 Popular use cases

Find TikTok influencers and micro-influencers

Search topics such as skincare, gaming, food, finance, travel, or fitness to find TikTok influencers, including smaller niche accounts, then export the results for campaign qualification.

TikTok creator discovery for brands and agencies

Run several product, competitor, or campaign terms together. Use search rank, account flags, and optional profile details to create a first-pass creator list.

TikTok user search by name or username

Run a TikTok user search with a display name, full username, or partial username to find matching public accounts and profile URLs.

Collect TikTok profile data

Enable enrichment to add available bios, bio links, counters, region, and full avatar URLs to discovery results.

Create a repeatable TikTok user search API workflow

Run searches through Apify's API or schedules, then send the dataset to spreadsheets, databases, dashboards, or other integrations.

🔄 A focused alternative to TikTok One and influencer platforms

This Actor is a focused alternative to TikTok One's Creator Marketplace and full influencer suites for public account discovery and structured data—not campaign management.

Existing option	Best fit	When to choose this Actor
TikTok app search	Interactive, one-query browsing	Run 1–10 keywords or names together and save ranked public profiles.
TikTok One / Creator Marketplace	Official filters, analytics, invitations, and projects	Discover creators by keyword or name without TikTok One access, then export or automate the dataset.
HypeAuditor, Modash, or Heepsy	Broader databases, analytics, vetting, outreach, tracking, or payments	Use a focused TikTok workflow billed per result, with API access and optional public-profile enrichment.

Choose this Actor for batch discovery, API or scheduled runs, and structured exports. Choose a full suite when you need audience demographics, engagement analysis, contact discovery, invitations, messaging, payments, or campaign reporting.

💵 Pricing

This Actor uses pay-per-result pricing with no start fee. Recent posts stay nested inside creator rows and are included at no extra charge, regardless of the number returned. Reading the expanded Recent videos view is also free of additional Actor charges. Enabling collection increases run time and proxy usage; existing creator and optional profile-enrichment charges remain:

Result	Price
Discovered creator	$0.005
Successful profile enrichment	+$0.002
Recent posts attached to a creator	Included — $0 extra

For example, 100 regular creator results cost $0.50. If all 100 profiles are successfully enriched, the total is $0.70. A blocked, private, or unavailable profile still keeps its discovery row but does not incur the enrichment charge.

🔗 Related TikTok Actors

Need a different kind of TikTok data? Continue your research with these Actors:

• 🎬 TikTok Video & Comments Scraper: Search public TikTok videos by keyword, collect engagement metrics, and optionally extract public comments for trend, product, competitor, and audience research.
• 🛍️ TikTok Shop Data Scraper: Extract TikTok Shop product data, including prices, reviews, seller information, and categories for e-commerce research and product discovery.

⚠️ Important limitations

• The Actor returns creator accounts, with recent public posts only when requested. It does not collect comments, follower lists, audience demographics, or private contact details.
• Search results come from TikTok's creator suggestions and are not a complete TikTok influencer database.
• The requested result count is a target, not a guarantee. A search stops when TikTok returns no new matching creators.
• The current search environment uses US routing and does not provide country, language, follower-range, engagement, or audience filters.
• Profile enrichment works only when a public profile is available. Unavailable profiles remain in the dataset with an appropriate status.
• Recent feeds use an unofficial public-web endpoint. Availability and full account-history coverage are not guaranteed. Private feeds, CAPTCHA, login automation, downloads, transcripts and niche classification are not included.
• Results can change as TikTok updates profiles, recommendations, and platform behavior.

❓ Frequently asked questions

How to find a TikTok creator?

Enter a creator name, username, niche, brand, or keyword in searchQueries, choose a target, and run the Actor. TikTok returns matching creator suggestions in ranked order. Enable enrichment for available bios and public account counters, then export the dataset to qualify the creators.

Can I search TikTok users by name?

Yes. The Actor works as a TikTok username search tool for public accounts. Search a full or partial display name or username to return available usernames, nicknames, profile URLs, TikTok IDs, and result ranks without providing TikTok credentials or accessing private account data.

Why are some results or profile fields missing?

TikTok may return fewer suggestions than requested or omit counters. The Actor requests additional deduplicated results where possible. Turn on enrichProfiles for more public details; private, blocked, or unavailable profiles remain in the dataset with an enrichment status.

Can I export the TikTok creator list?

Yes. Download the dataset from Apify Console in JSON, CSV, Excel, XML, or another supported format, or retrieve it through the Apify API.

🛡️ Responsible use

This Actor collects publicly available TikTok creator information. Use it lawfully, respect privacy and intellectual-property rights, follow applicable data-protection rules, and review TikTok's terms. Review the included licensing notices before redistributing the Actor image.