Website Email Scraper & Contact Finder: Phones & Socials

Answer first: this actor is a website email scraper and contact details scraper (an outreach contact finder) for link-building outreach and lead enrichment. Give it a list of website URLs or domains. For each domain it reads the homepage and the most likely contact, impressum, about, team, support and press pages, then returns emails, phone numbers (including WhatsApp click-to-chat numbers), the contact page URL and the company's LinkedIn, X (Twitter), Facebook, Instagram, YouTube and TikTok profiles, one row per domain. Junk is filtered out (example@ addresses, image@2x.png, Sentry/Wix tracking addresses, noreply). The first 5 domains of every run are free, then $0.004 per domain where contacts were found ($4 per 1,000). Domains with nothing found, dead sites and robots-blocked sites are never billed.

Quick facts
Best for	Extracting emails from a list of websites, link-building outreach lists, guest-post and PR prospecting, lead enrichment, CRM clean-up, finding a company's social profiles
Input	A list of domains or URLs (plausible.io, https://basecamp.com)
Output	One row per domain: primaryEmail, emails, phones, contactPageUrl, linkedin, x, facebook, instagram, youtube, tiktok, aboutPages, plus emailDetails (where each email was found) and phoneDetails (with a normalized digits form)
Price	First 5 domains free every run, then $0.004 per domain with contacts. Same price for 1 or 10 pages per domain. No start fee
Needs	No API keys, no proxies, no login. Public pages only
Not included	Email verification (SMTP), people search by job title, JavaScript-rendered pages, Google Maps

Try it now: the input is prefilled with 3 domains, so the first run is free.

Sample output (real run, 19 Oct 2026, 12 domains)

Unedited output of build 0.1.5, run 4lIeBJ6uvZbajg9M8: {"urls": ["plausible.io", "basecamp.com", "ahrefs.com", "usefathom.com", "ghost.org", "buffer.com", "joesbakery.com", "franklinbbq.com", "tartinebakery.com", "vapiano.de", "expired.badssl.com", "nonexistent-cygnix-xyz.com"], "maxPagesPerDomain": 5}. 35 pages fetched in 3.9 seconds of run time. 10 domains had contacts (5 free, 5 billed: $0.02), the expired-certificate test site was read but had no contacts (free), and the made-up domain came back unreachable (free). Platform cost of the whole run: $0.00029.

Download the full sample (all 12 rows with every field, public, no login): JSON · CSV · Excel

Domain	Best email	Phones	Contact page	Socials found
plausible.io	hello@plausible.io	-	plausible.io/contact (+ imprint)	LinkedIn, X
usefathom.com	support@usefathom.com	-	usefathom.com/about/contact	LinkedIn, X, YouTube
ghost.org	support@ghost.org	-	ghost.org/contact/ (+ help)	-
basecamp.com	jason@basecamp.com	-	basecamp.com/support	-
ahrefs.com	support@ahrefs.com (+ pr@ahrefs.com)	-	- (about + team pages read)	LinkedIn, X, Facebook, Instagram, YouTube
buffer.com	press@buffer.com	-	buffer.com/press	LinkedIn, X, Facebook, Instagram
joesbakery.com	contact@joesbakery.com (+ a Gmail address)	+15124720017, 7346651852	joesbakery.com/contact-us/	X, Facebook, Instagram
franklinbbq.com	hello@franklinbbq.com	512-653-1187	franklinbbq.com/contact	Facebook, Instagram
tartinebakery.com	info@tartinebakery.com (+ press@)	-	tartinebakery.com/press	Facebook, Instagram
vapiano.de	info@vapiano.de	05121 - 9127 - 0	vapiano.de/kontakt (+ impressum)	Facebook, Instagram, TikTok
expired.badssl.com	-	-	-	- (status ok (no contacts found), tlsWarning: invalid TLS certificate (CERT_HAS_EXPIRED); read anyway)
nonexistent-cygnix-xyz.com	-	-	-	- (status unreachable, error: ENOTFOUND)

Ahrefs and Tartine hide their addresses behind Cloudflare email protection; the actor decodes those. Buffer and Basecamp have no /contact page: the actor now also reads support and press pages, which is where press@buffer.com was found (build 0.1.3 returned no email for Buffer).

One row from the dataset (trimmed):

{
  "domain": "joesbakery.com",
  "status": "ok",
  "siteName": "Joe's Bakery & Coffee Shop - Mexican Food | Austin",
  "primaryEmail": "contact@joesbakery.com",
  "emails": ["contact@joesbakery.com", "basilicofood123@gmail.com"],
  "emailDetails": [
    { "email": "contact@joesbakery.com", "source": "mailto link", "page": "https://joesbakery.com/", "sameDomain": true },
    { "email": "basilicofood123@gmail.com", "source": "mailto link", "page": "https://joesbakery.com/", "sameDomain": false }
  ],
  "phones": ["+15124720017", "7346651852"],
  "phoneDetails": [
    { "phone": "+15124720017", "normalized": "+15124720017", "source": "schema.org", "page": "https://joesbakery.com/" },
    { "phone": "7346651852", "normalized": "7346651852", "source": "tel link", "page": "https://joesbakery.com/" }
  ],
  "contactPageUrl": "https://joesbakery.com/contact-us/",
  "aboutPages": ["https://joesbakery.com/about-us/"],
  "x": "https://x.com/joesbakeryatx",
  "facebook": "https://www.facebook.com/joesbakeryandcoffeeshop",
  "instagram": "https://www.instagram.com/joesbakeryeatx",
  "pagesCrawled": 3,
  "tlsWarning": null,
  "charged": false
}

What you get per domain

Field	What it is
primaryEmail	The best outreach address: on the site's own domain first, generic inboxes (hello@, info@, contact@, press@, partners@, kontakt@) before privacy@ / legal@ / dpo@ / DSA-authority / webmaster@ / jobs@ addresses
emails, emailDetails	Every unique email, lower-cased and de-duplicated, with the source (mailto link, page text, schema.org, cloudflare-protected), the page it was on and sameDomain (true for the domain you entered and the domain it redirects to, e.g. netcup.de -> netcup.com)
phones, phoneDetails	Numbers from tel: links, WhatsApp click-to-chat links (wa.me, api.whatsapp.com), schema.org telephone and text right after a "Phone / Tel / Telefon / Call us" label. The same number with and without country code is merged, including German-style +49 (0)9831 ... vs 0049 9831 .... phoneDetails[].normalized is the digits-only form (with + when the site gave a country code) for CRM import
contactPageUrl, contactPages	Contact, impressum / imprint / legal-notice, support / help and press pages actually fetched (English, German, French, Spanish, Italian, Portuguese paths). contactPageUrl is the real contact page when one exists, otherwise the impressum, support or press page
aboutPages	About, our story and team pages fetched
linkedin, x, facebook, instagram, youtube, tiktok	The profile that looks like the company (handle matches the domain or site name)
socialProfiles	All profile links found per network, e.g. founders' LinkedIn pages from a team page
status	ok, ok (no contacts found), unreachable, HTTP 404, blocked by robots.txt, error (with error for the reason, e.g. ENOTFOUND). Every domain you enter gets a row
tlsWarning, fallbackUrlUsed	Set when the site's TLS certificate is expired / self-signed / for another host and the page was read anyway, or when the www / non-www twin or http:// had to be used
charged	Whether this row was billed

A SUMMARY record in the key-value store has run totals (domains, with contacts, failed, billed, free).

What gets filtered out

Placeholder and junk emails never reach your list: example@, you@yourdomain.com, name@domain.com, john.doe@company.com, image file names that look like emails (logo@2x.png), Sentry and Wix tracking addresses (...@sentry.io, ...@sentry-next.wixpress.com), long hex hashes, noreply@ / donotreply@, and stray u003e / %20 prefixes from scripts (real addresses that start with digits, like 20north@..., are kept intact). Share buttons (twitter.com/intent, facebook.com/sharer), Instagram post URLs and TikTok video URLs are not counted as profiles. Phone placeholders like 0123456789 and US fictional 555-01xx numbers are dropped.

Input

{
  "urls": ["plausible.io", "https://basecamp.com", "ahrefs.com"],
  "maxPagesPerDomain": 5,
  "sameDomainEmailsOnly": false,
  "respectRobotsTxt": true,
  "maxConcurrency": 10
}

Field	Default	What it does
urls (required)		Domains or URLs, one per line. Duplicates and www / non-www twins are merged. startUrls ([{"url": ...}]) is accepted too
maxPagesPerDomain	5	Homepage + up to N-1 of the most likely contact / impressum / about / team pages (max 10). Price is per domain, not per page
sameDomainEmailsOnly	false	Keep only emails on the site's own domain
respectRobotsTxt	true	Skips pages robots.txt disallows for CygnixContactFinder
maxConcurrency	10	Domains in parallel. Pages within one domain are fetched one at a time
requestTimeoutSecs	20	Per page

What a run costs

You pay only for domains where at least one email, phone or social profile was found. The first 5 such domains in every run are free. No start fee.

Domains with contacts	You pay
5	$0 (free in every run)
100	$0.38
500	$1.98
1,000	$3.98
10,000	$39.98

Set Maximum cost per run in the run options and the actor stops cleanly when it is reached. It runs on 512 MB with plain HTTP (no browser, no proxy), so it is fast: the 10-domain sample above took under 8 seconds.

How it compares (live Apify Store prices, verified 19 Oct 2026, spot-checked 21 Oct 2026)

List prices on the Apify Free plan, read from the Apify Store API on 19 October 2026 (18:20 Riyadh time). Some actors charge less on higher Apify plans. "1,000 domains" assumes one run and about 5 pages per domain.

Actor	Price on the Free plan	Cost for 1,000 domains	Monthly users	Billed when nothing is found?
cygnix_labs/outreach-contact-finder (this)	$0.004 per domain with contacts, first 5 free every run, no start fee	$3.98	new	No
vdrmota/contact-info-scraper	$0.002 per page + $0.001 start per GB; paid lead / social add-ons	~$10 (5 pages); ~$2 homepage only	2,977	Yes (per page)
automation-lab/website-contact-finder	$0.035 per run + $0.001 per website returned; email verification $0.002	~$1.04	94	Yes
solidscrape/contact-details-scraper	$0.00125 per result + tiny start fee	~$1.25 (1 row per site)	56	Yes
dami_studio/contact-details-scraper	$0.0015 per site with contacts	~$1.50	3	No
datagleaner/website-contact-details-scraper	$0.002 per website with an email	up to $2	17	No
siftwright/contact-details-extractor	$0.002 per website scanned + tiny start fee	~$2	5	Yes (failed loads free)
poidata/contact-details-scraper	$0.002 per result + tiny start fee	~$2 (1 row per site)	54	Yes
jfaro19/website-contact-finder	$0.003 per website with an email or phone + tiny start fee	up to $3	13	No
khadinakbar/contact-details-scraper	$0.003 per domain + tiny start fee	~$3	27	Yes
emastra/website-contact-scraper	$0.004 per domain + tiny start fee	~$4	58	Yes
logiover/website-contact-scraper	$0.005 per result	~$5	37	Yes
thenetaji/website-email-scraper	$0.005 per result (Free plan)	~$5	78	Yes
jurassic_jove/website-email-extractor	$0.006 per result + tiny start fee	~$6	75	Yes
compass/contact-details-scraper-standby	$0.006 per website	~$6	1	Yes
delicious_zebu/contact-info-scraper	$0.01 per result	~$10	99	Yes
caprolok/website-email-phone-finder	$0.02 per result	~$20	81	Yes

Honest take. vdrmota's Contact Details Scraper is the proven leader (4.7 stars, 92 reviews) and does more: deep crawls, browser rendering, residential proxies and paid lead enrichment. It costs about $0.01 per domain at 5 pages, or about $0.002 if you only scrape the homepage. On big single runs several actors are cheaper per domain than us (automation-lab, solidscrape, dami_studio, datagleaner, siftwright, poidata, jfaro19, khadinakbar). Pick this actor when you want small and medium runs free or close to free (5 free domains in every run, nothing billed for empty or dead domains), homepage + contact + impressum + about + team + support + press pages for one flat price, a ranked best outreach email and the source page of every email and phone. code_crafter/leads-finder (2,012 users) is a different tool: it searches a B2B people database by job title rather than reading websites.

For AI agents / MCP

When an agent should call this actor: the user has a list of websites or domains and asks for their contact emails, phone numbers, contact page or social media profiles, wants to build an outreach or link-building list, or wants to enrich leads or a CRM export with emails and LinkedIn pages.

Connect the Apify MCP server with only this actor enabled (works in Claude, ChatGPT, Cursor, VS Code, Codex and any MCP client):

https://mcp.apify.com?tools=cygnix_labs/outreach-contact-finder

Or search for it with the default server: search-actors keywords such as contact details scraper, email finder from website, outreach contact finder, social links extractor.

Example call-actor input:

{
  "urls": ["plausible.io", "ghost.org", "franklinbbq.com"],
  "maxPagesPerDomain": 5
}

What to read from the result: each dataset row is one domain: primaryEmail (best outreach address), emails, phones, contactPageUrl, linkedin, x, facebook, instagram, youtube, tiktok and status. emailDetails[].page tells you where each address was published.

Cost control for agents: 5 domains are free in every run, then $0.004 per domain with contacts. Pass maxTotalChargeUsd in callOptions to hard-cap spend. This actor is pay-per-event, so it also works with Apify agentic payments (no Apify account needed for the agent).

Example prompt: "Find the contact email, contact page and LinkedIn for these 40 blogs so I can pitch a guest post."

FAQ

Where does it look? The page you give (or the homepage), then the best-matching links on that page: contact, impressum / imprint, about, team, support / help and press. Short paths win over blog posts that only mention "contact" in the slug. If no contact link exists it tries /contact and /contact-us (Shopify stores: /pages/contact; .de / .at / .ch sites: /kontakt, plus /impressum when no imprint link is found). Privacy, cookie, terms and careers pages are skipped even when they sit under /kontakt/. It stays on the same domain and its subdomains.

Does it verify emails? No SMTP or mailbox checks. Every email was published on the site itself; emailDetails shows the exact page and how it was found (mailto link, page text, schema.org, Cloudflare-protected). Run a verifier before cold email at scale.

Does it handle obfuscated emails? Cloudflare email protection is decoded, and simple name [at] domain [dot] com text is recognised. Emails drawn as images or loaded by JavaScript are not.

Does it render JavaScript? No. It reads the HTML the server sends, which keeps it fast and cheap. Most business sites put contact details in server HTML; single-page apps that load everything with JavaScript may return fewer results.

What about sites with an expired or self-signed certificate? Small-business sites often have one. The actor reads the public page anyway (read-only, robots.txt still honoured) and flags the row with tlsWarning. If the domain does not answer, it also tries the www / non-www twin and plain http://, and says which one worked in fallbackUrlUsed.

Does it find WhatsApp numbers? Yes, when the site links to wa.me/<number> or api.whatsapp.com/send?phone=<number>; they appear in phones with source whatsapp link.

Facebook pages without a username? Links like facebook.com/profile.php?id=1000... keep their id, so the link still opens the page.

Why was a domain not billed? It had no email, phone or social profile on the pages checked, was unreachable, returned an error, or robots.txt blocked the homepage. You also get 5 free domains in every run.

Is it legal? It only reads public pages, sends one request at a time per domain, identifies itself (CygnixContactFinder) and respects robots.txt by default. Emails can be personal data under GDPR and similar laws; you are responsible for how you use them (legitimate interest, opt-out, CAN-SPAM).

Can it find people by job title or company LinkedIn employees? No, it reads what the website publishes. For people search use a B2B leads database actor.

Does it work on Shopify stores? Yes. Shopify themes often hide the contact link in a JavaScript menu, so the actor also tries /pages/contact on Shopify sites (0.1.7: deathwishcoffee.com went from 0 to 1 email this way).

Does it work as an impressum scraper for German sites? Yes. Impressum / imprint pages are read for every domain, and on .de, .at and .ch sites /impressum and /kontakt are tried even when the homepage does not link them. Company email and phone usually sit there.

A mailto link lists several addresses? mailto:a@site.com,b@site.com gives both addresses (0.1.7).

Can I feed it Google Maps results? Yes: export the website column from a Google Maps scraper and paste it into urls.

Integrations

Run it through the Apify API, schedule it, or send results to Google Sheets, HubSpot, Make, Zapier, n8n, Clay or a webhook. Export as CSV or Excel for any outreach tool.

Changelog

• 0.1.7 (2026-10-21): Fixes from real-site tests: emails on the domain you entered were marked sameDomain: false (and dropped by sameDomainEmailsOnly) when the site redirects to another brand domain (netcup.de -> netcup.com, fathom.video -> fathom.ai); privacy-policy pages under /kontakt/ were picked as the contact page; +49 (0)... and 0049... copies of one number were not merged; mailto: links with several addresses were dropped; tracking-parameter URLs were crawled as contact pages; an unexpected error dropped the domain from the output (now an unbilled error row). New: Shopify /pages/contact and DACH /kontakt + /impressum guesses; DSA-authority, DPO and webmaster@ addresses rank below generic inboxes. No price change.
• 0.1.6 (2026-10-19): README: new real sample (run 4lIeBJ6uvZbajg9M8, build 0.1.5), comparison table re-verified 19 Oct 2026 with 13 rival actors, new FAQs. No price change.
• 0.1.5 (2026-10-19): Sites with an expired, self-signed or wrong-host TLS certificate are read instead of reported unreachable (flagged in tlsWarning); robots.txt is honoured on them too.
• 0.1.4 (2026-10-19): Fixes: emails starting with 20 or 3e were cut (20north@ became north@); Facebook profile.php?id= links lost their id; blog posts with "contact" in the URL could take the slot of the real contact page; a contact link that redirected to the homepage was read twice. New: WhatsApp numbers, support / help / press pages, phoneDetails[].normalized, www / non-www and http:// fallbacks (fallbackUrlUsed).
• 0.1.3 (2026-10-19): First public release (the sample above comes from build 0.1.2, same code; 0.1.3 adds this README). Domains or URLs in, emails + phones + contact/impressum/about/team pages + 6 social networks out. Cloudflare email decoding, junk filter, best-email ranking, phone de-duplication. $0.004 per domain with contacts, first 5 free every run. Rival prices verified 19 Oct 2026.

More tools by Cygnix Labs

• Website SEO Audit & Site Audit + AEO/GEO Checker
• Internal Link Audit & Orphan Page Finder
• AEO & GEO Audit + llms.txt Checker (URL list)
• Keyword Clustering + Content Briefs
• Search Intent Classifier
• Free local SEO MCP server: cygnix-seo-mcp (npx -y cygnix-seo-mcp)

Questions or a feature request? Open an issue on this actor's Issues tab. We answer within 24 hours.