YouTube Influencer Finder — Channel Search & Public Contacts

YouTube Influencer Finder searches public YouTube channels by keyword, filters subscriber counts and exports channels with a recognized public email, Instagram or X/Twitter contact. Use it for sponsorship research, creator shortlists and outreach planning. It does not reveal protected business emails or verify email deliverability.

What it searches

The Actor uses live HTTP requests to channel-only YouTube search results and public About pages; no browser or downstream Actor is started. There is no public provided-data or mock mode. The original query runs first, with search continuations where available.

When the original query falls short, keywordExpansion controls related discovery:

• auto prefers curated niche aliases, then live YouTube autocomplete suggestions, then generic creator/channel suffixes.
• broad mixes curated aliases and autocomplete suggestions, with additional generic suffixes available.
• off searches only the original query.

Related searches share the original keyword's output cap, subscriber filters and candidate budget. Channel IDs are deduplicated globally; the same channel is not returned again under a later keyword. Public profiles are cached within the run. Search matching is determined by YouTube, not an independent guarantee that a channel belongs to a particular niche.

Every saved row must have one recognized public email, Instagram or X/Twitter contact; there is no contact-filter-off input. A website-only link does not qualify. Parsing reads channel descriptions and public link values—it does not visit linked websites, test inboxes, confirm social ownership or bypass YouTube's protected business-email flow.

Quick start

Use this illustrative input in the Input tab:

{
  "keywords": ["AI tools"],
  "maxChannelsPerKeyword": 5,
  "minSubscribers": 1000,
  "keywordExpansion": "auto",
  "maxRelatedKeywords": 4
}

Start with a small result cap and inspect contact quality before expanding. A social handle can satisfy the contact requirement even when no email is found.

For API use, save the input as input.json:

curl -sS -X POST \
  "https://api.apify.com/v2/acts/rnlh9mbrUEjkkCtbt/runs" \
  -H "Authorization: Bearer ${APIFY_TOKEN}" \
  -H "Content-Type: application/json" \
  --data-binary @input.json

The response includes data.id and data.defaultDatasetId. Wait for completion and inspect status/logs before downloading the dataset as JSON, CSV or Excel. You do not supply a YouTube login or YouTube Data API key; the Apify API requires Apify authentication.

Input reference

All input fields are optional. There are no country or language filter inputs; discovery requests use English/US page parameters, and country is an output field when available.

Field	Default	Behavior
keywords	20 common niches	String array; up to 50 unique normalized keywords. Empty/omitted array uses built-in niches.
minSubscribers	0	Inclusive minimum; 0 disables it.
maxSubscribers	0	Inclusive maximum; 0 disables it. Must not be lower than an active minimum.
maxChannelsPerKeyword	30	1–200 qualifying output channels per original keyword, not raw candidates or separate caps per related query.
keywordExpansion	auto	auto, broad, off; expansion is conditional on a result shortfall.
maxRelatedKeywords	4	1–10 related searches per original keyword.
maxConcurrency	12	1–25 parallel profile requests; search queries/continuations are processed in sequence.
maxRequestRetries	1	0–3 retries for supported temporary request failures.
requestTimeoutSecs	30	10–120 seconds per HTTP request, not the whole run.
proxyConfiguration	disabled	Direct HTTP by default; selected proxies are used only when explicitly enabled. Proxy costs may apply.

Known out-of-range search counts are excluded before About-page requests. When a subscriber bound is active, channels with unavailable counts are excluded too. Counts parsed from abbreviated text such as 12.3K subscribers are approximations of the displayed source count, not exact private analytics.

Output and billing unit

Each dataset row is one retained channel, including:

• channelName, handle, channelId, channelUrl, avatar, description, keywords, country, subscriberCount.
• email, emails, instagram, twitter, links, primaryContactType, primaryContact. Primary contact preference is email, Instagram, then X/Twitter.
• keyword, matchedKeywords, matchedQuery, relatedKeyword, expansionSource, matchedSearchText and ISO scrapedAt.

matchedKeywords currently contains the keyword for which the row was emitted; it is not a complete list of every later query that could match that channel. Counts, country and avatar can be unavailable. scrapedAt is processing time, not proof of the source's update time.

Filtered, contact-bearing and deduplicated records are passed to Actor.pushData, which uses the apify-default-dataset-item event. Rejected candidates do not incur a result-row event. The event budget and Free-plan limit can truncate writes. No saved rows means the run fails; some saved rows do not establish that every keyword succeeded or the search was complete.

Current pricing

Plan	Per channel row	Per 1,000 rows
FREE	$0.0225	$22.50
BRONZE	$0.0045	$4.50
SILVER	$0.0040	$4.00
GOLD, PLATINUM, DIAMOND	$0.0035	$3.50

Actor Start: $0.00005 per event, one event per GB of allocated memory with a minimum of one event per run. Startup is not a fixed total at every memory setting and can be charged even for empty/failed runs.

Apify platform usage is billed separately to the running user. Compute, transfer, storage and explicitly selected proxy use can incur costs even when no contacts are saved. Row-fee examples exclude these costs. Consult the Pricing tab for current rates.

Free-plan policy: at most 50 dataset rows globally per run, across all keywords. This is a developer-set limit. Paid plans have no developer-set 50-row limit, but input, source, safety and budget limits remain. FREE does not mean zero-priced results.

Limits and responsible use

Discovery has a calculated budget of at most 4,000 unique candidate profiles per original keyword and at most 15 search pages per query. It can stop at a page/candidate/query budget, missing continuation data, source exhaustion or a charge limit—before reaching your requested result count. Narrow subscriber ranges and globally overlapping results reduce output.

YouTube layouts, consent pages, blocking and hidden or deleted channels can prevent extraction. Public contacts can be missing, stale or misparsed; validate them before outreach. This is neither a complete channel directory nor a guarantee of ownership, niche relevance or delivery. Follow privacy, anti-spam and platform rules.

FAQ

Does every result have an email? No. Every row has a recognized supported contact, which may be Instagram or X/Twitter only. No email is delivery-tested.

Can it obtain the email behind YouTube's business-email button? No. It only parses exposed page text/link data; it does not bypass CAPTCHA or restricted access.

Will maxChannelsPerKeyword: 100 return 100 rows? It is an upper bound. Availability, filters, overlap, bounded discovery and billing limits may yield fewer rows.

Complementary Actors

• YouTube Scraper — collect videos, Shorts and channel content metadata.
• YouTube Comments Scraper — inspect public discussion on selected videos.