Contact Details Scraper - Website Emails, Phones & Socials API

This contact details scraper is a website email extractor, phone number finder and social media link scraper for B2B lead generation: give it a list of company websites or domains and get one clean row per website with every business email, phone number (international format), LinkedIn company page, Facebook, Instagram, X/Twitter, YouTube, TikTok, Pinterest, GitHub, WhatsApp and Telegram link, the company name, postal address (from structured data), logo and the URL of the contact page.

It reads the homepage plus the most contact-like pages it can find (contact, imprint / Impressum, about, support, locations) over plain HTTP, with no headless browser, so 1,000 websites finish in minutes. Unlike per-page contact scrapers, you pay per website, no matter how many pages it reads.

What you get

One dataset item per website:

• Company: companyName (from schema.org data, og:site_name or the page title), title, description, language, country (from the domain ending), logo, address (from schema.org PostalAddress when the site publishes it).
• Emails: email (the best one: same domain and role inboxes like info@ / sales@ first), emails (all, ranked), emailSources (where each was found: mailto, json-ld, cloudflare, text plus the page path). Cloudflare-protected addresses are decoded, and info [at] acme [dot] com style obfuscation is read.
• Phones: phone, phones (international format, e.g. +49 9831 5050) and phonesE164 (+4998315050), validated with libphonenumber. Clickable tel: links and structured data come first.
• Social profiles: linkedin, facebook, instagram, twitter, youtube, tiktok, pinterest, github, threads, whatsapp, telegram, discord, yelp, trustpilot (the site-wide profile, preferring links in the homepage header/footer), plus socialProfiles with every profile found per network. Share buttons, posts and personal LinkedIn profiles are ignored.
• Crawl info: contactPageUrl, pagesCrawled, pageUrls, finalUrl (after redirects), status (ok or failed) and scrapedAt.

Websites that cannot be loaded (domain does not exist, server down, strict bot protection) get a row with status: "failed" and an error. Those rows are free. A RUN_SUMMARY record lists totals and fill rates.

Sample output

Real rows (one row per website; hetzner and apify from sample-output.json, sennheiser and kinsta from a 67-site cloud run):

domain	companyName	email	phone	linkedin	pagesCrawled
hetzner.com	Hetzner	info@hetzner.com	+49 9831 5050	https://www.linkedin.com/company/hetzner-online	6
apify.com	Apify Technologies s.r.o.	hello@apify.com	-	https://linkedin.com/company/apify	6
sennheiser.com	Sennheiser	contact@sennheiser.com	+49 5130 6000	https://www.linkedin.com/company/sennheiser	6
kinsta.com	Kinsta Inc.	contact@kinsta.com	+1 888 610 2915	https://www.linkedin.com/company/kinsta	6

Use cases

• Enrich a lead list: you have company domains from Google Maps, LinkedIn, a trade-show list or a CRM export; add emails, phones and social links.
• Sales prospecting: find the general inbox, sales line and LinkedIn company page for every account before outreach.
• Agency and SEO outreach: collect contact emails and social handles for a list of sites.
• Data hygiene: re-check that the emails and phones in your CRM are still the ones on the website.
• AI agents: one call turns a list of domains into structured, deduplicated contact data.

Pricing

$1.20 per 1,000 websites (pay per result; failed requests and failed websites are never charged). One result = one website, however many pages were read. Per-page contact scrapers charge for every page they open, about 6 per site: the most-used one bills $2.00 per 1,000 pages (about $12 per 1,000 sites) plus extra events per lead and social profile, and the pay-per-website alternatives in the top results charge $1.25-$20.00 per 1,000.

You scrape	You pay
100 websites	$0.12
1,000 websites	$1.20
10,000 websites	$12.00
100,000 websites	$120.00

Plus a $0.00005 start fee per run. The Apify free plan's $5 monthly credit covers about 4,150 websites. The prefilled input (apify.com and hetzner.com, up to 6 pages each) returns 2 rows in about 15 seconds and costs about $0.0025.

Why this one

• The lowest per-website price in the top results: $1.20 per 1,000 sites, against $1.25-$20.00 for the other pay-per-website scrapers and about $12 for per-page billing.
• One flat price per website: up to 30 pages per site (maxPagesPerWebsite) for the same $0.0012, so deeper crawls do not multiply the bill.
• Failed websites are free: dead domains and strictly protected sites get a status: "failed" row with the error and cost nothing.
• Emails ranked, not just dumped: same-domain role inboxes (info@, sales@) first, with emailSources showing where each was found; Cloudflare-protected and [at] / [dot] addresses are decoded.
• Phones validated with libphonenumber, in international and E.164 format.
• Company social profiles only: site-wide LinkedIn, Facebook, Instagram, X, YouTube, TikTok and 8 more networks; share buttons and personal profiles are skipped.

FAQ

Is it legal to scrape contact details from websites? The actor only reads public pages and the contact details a business publishes about itself (general inboxes, switchboard numbers, company social pages). It skips personal LinkedIn profiles and does not follow team or people pages. Emails can still be personal data under GDPR and similar laws, so use the results for legitimate business purposes and follow anti-spam rules. This is not legal advice.

Does it need a login, API key or special proxy? No login and no API key. The default Apify Proxy (datacenter) works for the vast majority of company websites. About 5% of sites sit behind strict bot protection (Cloudflare, Akamai); they come back as status: "failed" and cost nothing.

How fast is it and how many websites can I send? It reads plain HTML with no browser and 20 parallel requests by default, so 1,000 websites finish in minutes. You can paste tens of thousands of domains; each becomes exactly one row, deduplicated by domain.

Why does a website have no email? Many sites only offer a contact form. The actor reports what the site publishes; it never guesses addresses. Single-page apps that render everything in the browser may also return fewer contacts.

Can I schedule it or use it via API? Yes: Apify Schedules, the REST API and JavaScript / Python clients, Make, Zapier, n8n, webhooks, or the Apify MCP server for AI agents.

Can I crawl more pages per site? Yes, up to 30 with maxPagesPerWebsite. The price stays per website.

How to use

1. Paste your Websites or domains, one per line: stripe.com, www.thomann.de or https://www.joesstonecrab.com. Lists exported from Google Maps, a CRM, a directory or a spreadsheet work as they are.
2. Optionally change Max pages per website (default 6; 1 = homepage only), set a Default phone country for numbers written without a prefix, or tick Only emails on the website's own domain.
3. Click Start. Download the results as CSV, Excel, JSON or HTML, or send them to your CRM, Google Sheets, Make, Zapier, n8n or an AI agent through the Apify API / MCP.

Input example

{
    "websites": ["apify.com", "hetzner.com", "https://www.joesstonecrab.com", "thomann.de"],
    "maxPagesPerWebsite": 6,
    "scanTextForPhones": true,
    "onlySameDomainEmails": false
}

Output example

Real row from a cloud run:

{
    "website": "hetzner.com",
    "domain": "hetzner.com",
    "finalUrl": "https://www.hetzner.com/",
    "status": "ok",
    "companyName": "Hetzner",
    "title": "Dedicated Servers, Cloud & Hosting from Germany | Hetzner",
    "language": "en-US",
    "email": "info@hetzner.com",
    "emails": ["info@hetzner.com", "dsa-authorities@hetzner.com", "tco@hetzner.com"],
    "emailSources": {
        "info@hetzner.com": "mailto @ /legal/legal-notice/",
        "dsa-authorities@hetzner.com": "mailto @ /legal/legal-notice/",
        "tco@hetzner.com": "mailto @ /legal/legal-notice/"
    },
    "phone": "+49 9831 5050",
    "phones": ["+49 9831 5050", "+49 9831 5053"],
    "phonesE164": ["+4998315050", "+4998315053"],
    "linkedin": "https://www.linkedin.com/company/hetzner-online",
    "facebook": "https://www.facebook.com/hetzner.de",
    "instagram": "https://www.instagram.com/hetzner.online",
    "twitter": "https://x.com/Hetzner_Online",
    "youtube": "https://www.youtube.com/user/HetznerOnline",
    "tiktok": null,
    "contactPageUrl": "https://www.hetzner.com/legal/legal-notice/",
    "pagesCrawled": 6,
    "scrapedAt": "2026-10-09T17:27:53.266Z"
}

On a 67-website test list (SaaS, e-commerce, restaurants, German retailers) 63 loaded; of those 60% had an email, 33% a phone, 87% at least one social profile, 83% a contact page and 25% a full postal address.

API example

Python (pip install apify-client):

from apify_client import ApifyClient

client = ApifyClient("<YOUR_APIFY_TOKEN>")
run = client.actor("rel8ble/website-contact-extractor").call(run_input={"websites":["kinsta.com","teamviewer.com","sennheiser.com"],"maxPagesPerWebsite":6})
for item in client.dataset(run["defaultDatasetId"]).iterate_items():
    print(item["domain"], item.get("email"), item.get("phone"), item.get("linkedin"))

One HTTP call that runs the actor and returns the rows (curl):

curl -X POST "https://api.apify.com/v2/acts/rel8ble~website-contact-extractor/run-sync-get-dataset-items?token=<YOUR_APIFY_TOKEN>" \
  -H "Content-Type: application/json" \
  -d '{"websites":["kinsta.com","teamviewer.com","sennheiser.com"],"maxPagesPerWebsite":6}'

Integrations

• API: start runs and read datasets over the Apify REST API or the JavaScript / Python clients.
• MCP / AI agents: available through the Apify MCP server; every input and output field is documented.
• No-code: Make, Zapier, n8n, Google Sheets, HubSpot, Airtable, Slack, webhooks.
• Chaining: feed it the website column of a Google Maps, directory or YC companies scraper.

How it works (for developers)

For each domain the homepage is fetched (falling back to www. / http:// variants), then same-site links are scored by contact likelihood in nine languages (contact, impressum, mentions-legales, about, support, locations...) and the best ones are fetched up to the page limit; if the homepage links are injected by JavaScript, /contact, /contact-us, /about and /impressum are probed. Every page is parsed with Cheerio for mailto: and tel: links, Cloudflare data-cfemail blocks, schema.org Organization / LocalBusiness JSON-LD (email, telephone, address, sameAs, logo) and social links; visible text is scanned for plain and obfuscated emails and, on contact-type pages, for phone numbers that libphonenumber validates as full-length numbers. Results are merged per domain and pushed once.

npm install
npm test             # extraction tests on synthetic and saved real pages
node src/main.js     # input in storage/key_value_stores/default/INPUT.json