Business Contact & Lead Finder

Give it company websites. Get back contactable people.

What it returns

Named decision-makers with job titles — not just an info@ address. On a real run: Bryn Richards [CEO], Peter Morris [Founder], Adam Kudryl [Director]. With a name and a title you can infer the email pattern; with info@ you cannot. Many SMB sites list their whole team while hiding emails behind a contact form, so this recovers leads an email-only scraper reports as nothing — one accountancy firm returned 41 named staff and zero emails.

The filtering is the product

A plain email regex over a homepage returns noreply@, Sentry ingest keys, logo@2x.png, Facebook's tracking pixel and the web agency's own address. A lead list you have to clean by hand is worth very little. This one:

• Separates named humans from generic inboxes. sarah.jones@acme.com is a lead. info@acme.com is a switchboard. They are returned in different fields because they are worth different amounts.
• Flags off-domain emails rather than hiding them — an address on another domain is usually the agency or a partner, not your prospect.
• Reads the site's own navigation instead of guessing URLs. Guessed paths return styled 404s that still parse as pages — /our-team looked like a successful scrape and was a not-found. Discovered links are the real pages, and this roughly tripled the yield.
• Trusts mailto: and tel: links first, because a human marked those up.
• Refuses obvious junk. An early version returned 2015-2026 — a copyright year range — as a phone number. Now a match needs a country code, parentheses or real separators, and 9-15 digits.

Output

{
  "website": "https://hjsolicitors.co.uk", "ok": true, "leadScore": 54,
  "peopleCount": 13,
  "people": [{ "name": "Bryn Richards", "title": "CEO" },
             { "name": "Toby Harper", "title": "Founder" }],
  "personalEmails": [], "roleEmails": ["enquiries@harperjames.co.uk"],
  "offDomainEmails": ["enquiries@harperjames.co.uk"],
  "phones": ["0800 689 1700"],
  "linkedinCompany": "linkedin.com/company/harper-james-solicitors",
  "socials": ["facebook.com/hjsolicitors", "linkedin.com/company/harper-james-solicitors"]
}

leadScore (0-100) weights named people and personal emails highest, then role inboxes, then a phone and a LinkedIn company page.

A site that could not be fetched is returned with ok: false and the reason — never silently dropped, because a missing row reads as "this company has no contact details" when the truth is the site did not load.

Works outside English

Contact-page names and job titles are matched in English, Spanish, German, French, Italian, Portuguese and Dutch — kontakt, impressum, quiénes-somos, equipo, chi-siamo, and titles like Socio Director or Geschäftsführerin. The English-only first version found zero people on every Spanish site tested.

Site-builder social accounts are also filtered: a Valencia dental clinic was reporting linkedin.com/company/wix-com as its own page. Shipping a vendor's profile as the lead's is worse than shipping nothing, because it looks like data.

Honest expectations

Yield depends on the sector. Professional services, trades and local businesses publish staff pages and contact details. Large consumer brands hide behind contact forms and will often return only a phone, socials and whatever names are public. Page headings are filtered out — an early version reported "Who We Help" as a CEO. Use onlyWithEmail if you only want to pay for records that carry an address.

Pairs with

Feed it the website column from any company or directory scraper — this is the enrichment step that turns a list of names into a list you can actually contact.