Business Website Contact Finder (Generic Emails Only) avatar

Business Website Contact Finder (Generic Emails Only)

Pricing

$1.50 / 1,000 websites

Go to Apify Store
Business Website Contact Finder (Generic Emails Only)

Business Website Contact Finder (Generic Emails Only)

Find generic business contacts on a company website: info@, sales@, phone, address and official social links. Personal-name e-mails are filtered out and never written. Respects robots.txt and limits pages per site.

Pricing

$1.50 / 1,000 websites

Rating

0.0

(0)

Developer

Open Data Desk

Open Data Desk

Maintained by Community

Actor stats

0

Bookmarked

2

Total users

1

Monthly active users

2 days ago

Last modified

Features

Limited permissions

Share

Find the public business contacts on a company website: generic role addresses such as info@ and sales@, phone numbers, an address, and official social links. Personal-name e-mails are filtered out and are never written to the results.

This is meant for sales teams in the EU who want a company contact, not a list of people. It also works as a small block in n8n, Make, Zapier or an AI agent.

What data do you get?

One row per website.

FieldWhat it is
website, finalUrlThe site you asked for and the first page that responded
emailsGeneric role addresses only, with the page they were found on
phonesPhone numbers from tel: links and page text
addressesPostal addresses from JSON-LD when the site publishes them
socialCompany page links (LinkedIn company pages, and similar). Personal profile URLs such as LinkedIn /in/ are left out
pagesCrawled, pagesSkippedRobotsHow many pages were read, and how many were blocked by robots.txt
emailsSkippedPersonalHow many addresses were dropped as personal, consumer-mail or not generic. The addresses themselves are not stored
statusok, no_contacts, blocked_robots, failed, invalid_url or skipped_limit

Example output

One row from a test run on example.com, which has no contacts. The row is still returned.

{
"inputUrl": "https://www.example.com",
"website": "https://www.example.com/",
"finalUrl": "https://www.example.com/",
"emails": [],
"phones": [],
"addresses": [],
"social": [],
"pagesCrawled": 1,
"pagesSkippedRobots": 0,
"emailsSkippedPersonal": 0,
"skipReasons": {
"personal_name": 0,
"consumer_domain": 0,
"not_generic": 0,
"invalid": 0
},
"status": "no_contacts",
"error": null,
"scrapedAt": "2026-11-16T14:55:23.302Z"
}

How personal e-mails are filtered

An address is kept only if all of this is true:

  • The local part is a known generic role, or a short combination of roles, such as info, kontakt, sales, customer.service or hello.
  • It does not look like a person's name (jane.doe, j.smith, and similar).
  • The domain is not a consumer mailbox such as Gmail, Outlook, Yahoo, iCloud, Proton or Seznam.

Addresses that fail the test are counted in emailsSkippedPersonal and in skipReasons. They are not written out, even in the log.

This is deliberately strict. A first name on a company domain, such as anders@firma.dk, is not returned. It is better to miss a contact than to return a personal address.

Nothing is silently dropped

Every website you type in is written back. Sites that block the crawler, fail to load, or have no generic contact still get a row with a status. If you set Max sites lower than the size of your list, the remaining sites are still written with status skipped_limit.

How to use it

  1. Paste company websites in Business websites.
  2. Leave Max pages per site at 6, or raise it a little if a site hides the contact page.
  3. Click Start and download the result.

The Actor starts on the homepage, tries common contact paths (/contact, /kontakt, /impressum, /about, and similar), and follows a few same-site links that look like contact pages.

Limits and robots.txt

  • robots.txt is read first. Disallowed paths are skipped. If the file blocks this crawler from the whole site, you get status blocked_robots.
  • A short delay is used between pages. If robots.txt sets a crawl delay, it is honoured up to five seconds.
  • Only HTML on the same host is read. No search engines, no login areas, no documents.
  • Addresses are taken from structured data when it is there. Free-text footer addresses are easy to get wrong, so they are not guessed.

How much does it cost?

You pay for each website that is written, including failed, blocked and skipped ones. The price per 1,000 sites is shown at the top of this page, and you can set a maximum cost per run.

The Actor only reads the public pages you point it at, and only after checking robots.txt. E-mail addresses can be personal data. This Actor is built to return generic company role addresses only, and not to keep the ones it filters out. You are still responsible for how you use the contacts, including marketing rules that apply in your country.

Issues and requests

If a generic address is missed, or a personal one slips through, please open an issue on the Issues tab. You can send the website and the kind of address, without the person's name.