Contact Details Extractor

Website URLs in, structured contact details out.

Which contact-details Actor should I use?

Use this one when you already have the page URLs (a contact page, a team page, a company homepage) and want emails, phones, and social links back as one JSON record per URL at $0.0015 per page. Use a crawler-style contact scraper instead if you need the whole site crawled, JavaScript-rendered pages, or postal addresses.

Input	pageUrls (one page each; no crawl)
Output per URL	emails[] (≤50), phones[] (≤50, North American format), social{linkedin,x,facebook,instagram,youtube}, contactPages[] (≤20), status
Pricing	$0.0015 per page processed ($1.50 / 1,000). Error records are not billed.
Not supported	JavaScript-rendered contacts, contacts in images, postal addresses, contact forms, international phone formats
Call it	REST skilled_glee~contact-details-extractor · MCP https://mcp.apify.com/?tools=skilled_glee/contact-details-extractor

Jobs + contacts in two calls

1. Run LinkedIn Jobs for your keywords → company names and job URLs.
2. Pass each company's contact or about page URL to this Actor → emails, phones, socials. Both return one record per item and bill only successful records.

What it does

Fetches each page URL you give it (one page per URL; it does not crawl the site) and extracts, from that page's HTML: email addresses (including mailto: links), North American–format phone numbers, social profile links (LinkedIn, X/Twitter, Facebook, Instagram, YouTube), and links on the page whose URL contains contact, about, team, or support. Plain HTTP fetch with pattern extraction; no AI, no browser.

It does not extract physical/postal addresses, and it does not detect or submit contact forms.

Input

{
  "pageUrls": ["https://example.com/contact"],
  "useResidentialProxy": false
}

• pageUrls (required): pages to fetch. If empty, the Actor falls back to https://www.example.com and charges for it normally, so always pass the URLs you want.
• useResidentialProxy (optional, default false): route fetches through Apify residential proxy for bot-walled targets.

Output

One dataset record per input URL:

{
  "url": "https://example.com/contact",
  "finalUrl": "https://example.com/contact",
  "emails": ["info@example.com"],
  "phones": ["+1-555-010-0100"],
  "social": {
    "linkedin": ["https://www.linkedin.com/company/example"],
    "x": ["https://twitter.com/example"]
  },
  "contactPages": ["https://example.com/about"],
  "linksFound": 42,
  "status": "ok"
}

• emails and phones: up to 50 each, de-duplicated in page order. Matches are pattern-based, so review them before use; phone matching targets 10-digit North American formats and can miss international numbers.
• social: only networks actually found appear as keys, up to 20 links each.
• contactPages: up to 20 links on the page whose URL contains contact, about, team, or support. These pages are listed, not fetched.

A URL that cannot be fetched produces {"url": "...", "status": "error", "error": "..."}. The key-value store record run-summary holds requested, succeeded, failed, billable_units, ppe_event, and useResidentialProxy.

Use cases

• Lead generation pipelines
• Business directory building
• Contact database enrichment
• Sales prospecting automation

Pricing

$0.0015 per page processed ($1.50 per 1,000 pages)

Charged once per status: "ok" record, meaning the page was fetched and parsed. Pages with no contact details found are still billed because processing occurred. Error records (unreachable page, HTTP error, invalid URL) are not billed.

Limitations

• Static HTML only; contact details rendered by JavaScript or shown in images are not extracted
• One page per URL; no site crawling
• No physical address or contact-form extraction
• Does not read or apply robots.txt; only submit URLs you are permitted to fetch
• Reads at most the first 2 MB of each page

Failure behavior

Each page is processed independently. Unreachable pages return explicit error records and the batch continues. Transient HTTP errors (429/5xx) and network errors are retried up to 3 times.