Contact Email, Phone & Social Extractor

Give this actor one or more company websites and it returns the contact details those sites publish: email addresses, phone numbers and links to their social profiles. Everything is deduplicated and grouped into one clean row per website, ready for a spreadsheet, a review list or your CRM. It is made for sales, recruiting and marketing people who need contact data without copying it by hand, and you do not need to write any code to use it.

What you get

One row per website, even when you give it several links from the same site. Each row contains:

• Email addresses the site shows, including ones written as name [at] example [dot] com and ones a site hides behind its spam protection.
• Phone numbers, reduced to digits. Numbers with a country code keep the leading +.
• Social profiles on LinkedIn, X (Twitter), Facebook, Instagram, YouTube, TikTok, GitHub, Pinterest, Threads and Telegram. Only real profiles: share buttons, single posts and videos are left out.
• Where each email and phone was found, so you can check the source page yourself.
• The pages that were visited, without tracking parameters.
• A list of pages that could not be opened, with the reason.
• The date and time the row was made.

Obvious junk is filtered out, such as noreply@ addresses, addresses at well-known placeholder domains and image file names.

How to use

1. Open the actor in Apify Console and go to the Input tab. The form already contains one example address; replace it with yours.

2. Fill in the form. Start URLs and Domains decide what gets visited, so at least one of them must have something in it.

   • Start URLs — full web addresses, including https://, for example https://www.example.com/contact. Add as many as you need.
   • Domains — site names only, for example example.com. Useful when you do not know the exact contact page. Several are allowed.
   • Max link depth — how many links away from your starting page the actor may go. 0 visits only the pages you listed, 1 also visits pages linked from them, and so on. Allowed: 0 to 5. Empty means 2.
   • Max pages per domain — the most pages it will try to open on one website, counting the starting page. Allowed: 1 to 100. Empty means 10. Use 1 to read only the page you gave it.
   • Validate email domains — off by default. Turn it on to check that the part of each email after the @ is set up to receive mail. Each email then gets a status: has_mx (can receive mail), no_mx (cannot) or unknown (could not tell). Empty or off means no check, and the status is null.

   You can fill in both Start URLs and Domains; the two lists are combined and duplicates are removed.

   If Start URLs and Domains are both empty, the run stops right away with a message and you get no data. A mistyped address also stops the run, and the message names the address that needs fixing.

3. Press Start.

4. Wait for the run to finish. The log shows pages as they are read. Sites with many pages take longer; the page limit keeps a single site short.

5. Open the Output tab (the dataset). Each row is one website. Review the emails and phones, then use Download to save the rows as JSON, CSV or Excel.

6. If you prefer, the same rows can be fetched through the Apify API or passed to another tool from the Integrations tab, without any code.

Example

Input

{
  "startUrls": [{ "url": "https://www.apify.com/contact" }],
  "domains": [],
  "maxDepthLinks": 0,
  "maxPagesPerDomain": 1,
  "validateEmails": true
}

Output

One row of the output. Empty networks are shown here because they are always part of a row.

{
  "domain": "apify.com",
  "startUrl": "https://apify.com/contact",
  "crawledAt": "2026-10-27T05:57:48.611Z",
  "pagesCrawled": 1,
  "pages": [
    "https://apify.com/contact"
  ],
  "emails": [
    {
      "address": "hello@apify.com",
      "sourceUrls": [
        "https://apify.com/contact"
      ],
      "mxStatus": "has_mx"
    },
    {
      "address": "support@apify.com",
      "sourceUrls": [
        "https://apify.com/contact"
      ],
      "mxStatus": "has_mx"
    }
  ],
  "phones": [],
  "socialLinks": {
    "linkedin": [
      "https://linkedin.com/company/apify"
    ],
    "twitter": [
      "https://x.com/apify"
    ],
    "facebook": [],
    "instagram": [],
    "youtube": [
      "https://youtube.com/apify"
    ],
    "tiktok": [
      "https://tiktok.com/@apifyoffice",
      "https://tiktok.com/@apifytech"
    ],
    "github": [
      "https://github.com/apify"
    ],
    "pinterest": [],
    "threads": [],
    "telegram": []
  },
  "errors": []
}

This row comes from a real run of the actor.

If a page could not be opened, it appears in errors with the page address and a short reason.

Tips and limits

• Only what is visible on the page. Contact details that appear only after pressing a button, inside a form, or inside an image are missed. Pages whose text is added later can also come up empty.
• Some sites refuse automated visits. Those pages are listed under errors and the run continues. If every page of a site is refused, you still get a row, but its lists may be empty.
• One website at a time. Links to other sites and to subdomains are not followed. shop.example.com and example.com are two separate websites, each with its own row.
• Some pages are skipped. Login, account, cart, checkout, admin and search pages are not opened, and each site's rules for automated visits (robots.txt) are respected.
• Runs are gentle, so they can be slow. Pages are read a few at a time. Each run stops automatically after one hour by default; for a long list, split it into smaller runs or raise the limit in the run options.
• Phone numbers are a best guess. Numbers wrapped across lines, hidden inside codes, or written without separators may be missed or misread.
• The email check is not a mailbox test. A has_mx result means the domain is set up to receive mail. It does not prove the address exists, and no message is ever sent.
• Published sample addresses are kept. If a site itself publishes a demo address, it is collected like any other address.
• Set a realistic page limit. A low limit keeps runs quick and polite; a high one finds more, especially on large sites.

Use it responsibly

• The actor reports contact details that websites publish openly, and these can include personal data such as personal mailboxes and direct phone numbers.
• If you save or use the results, you are responsible for following privacy law, for example GDPR/UK GDPR and CCPA: have a lawful basis, collect only what you need, keep it only as long as needed, and honor deletion requests.
• Follow anti-spam law too, for example CAN-SPAM and the ePrivacy rules: contact only people you may contact, honor opt-outs, and do not send unsolicited bulk email.
• Visit only what you are allowed to visit. Do not use the actor to get past logins, paywalls or other protections, and do not use it to test whether a mailbox exists.
• The sample row in this README comes from a real run of the actor.

FAQ

Do I need to know how to code? No. Fill in the form, press Start, and download the rows.

How many websites can I do in one run? As many as you paste. Time grows with every website, and the default run limit is one hour, so split very long lists into a few runs.

Why does a row have no emails? The site may not publish any, may show them only inside a form or after pressing a button, or may have refused the visit. Check the pages and errors lists to see what happened.

Does “can receive mail” mean the address is real? No. It only means the domain is set up to receive email. The actor never contacts a mail server, so it cannot confirm that a particular mailbox exists.

Why did two of my links give one row? Because they belong to the same website: results are merged per site, and www. is ignored. Use two different sites, or two different subdomains, to get two rows.