Paid email finders mostly do one thing you can do yourself: work out the pattern a company uses for addresses, then apply it to a name. For small companies you can often get there in five minutes with a terminal and the company's own website.
This is the process I use, in the order I use it, including the point where I stop.
Step 1: Does the domain accept email at all?
Before guessing anything, check that the domain has mail servers. If it has no MX records, every address you guess will bounce.
With dig:
dig +short MX acme-plumbing.co.uk # 0 acmeplumbing-co-uk.mail.protection.outlook.com. With nslookup (works on Windows too):
nslookup -type=mx acme-plumbing.co.uk An empty answer means no mail. Some domains have no MX but do have an A record, and in theory mail can be delivered to that; in practice I treat no MX as "don't guess".
For a list of domains, Python with dnspython (pip install dnspython):
import dns.resolver def mx_hosts(domain): try: answers = dns.resolver.resolve(domain, "MX", lifetime=5) return sorted((r.preference, str(r.exchange).rstrip(".")) for r in answers) except (dns.resolver.NXDOMAIN, dns.resolver.NoAnswer, dns.resolver.NoNameservers, dns.exception.Timeout): return [] for d in ["acme-plumbing.co.uk", "example.com"]: print(d, mx_hosts(d) or "no MX") The MX host also tells you something. *.mail.protection.outlook.com is Microsoft 365, aspmx.l.google.com is Google Workspace. Both are real mailboxes run by the company. A domain pointing to a forwarding service or a web host's default mail is more likely to be a small setup where firstname@ or even just info@ is all there is.
Step 2: Find one real address
One published personal address tells you the format for everyone. Look in this order:
- The company's own site. Team page, contact page, job ads ("send your CV to anna@…"), the privacy policy, press page. Search the page source for
@yourdomainsince some addresses only appear inmailto:links. - PDFs on the site. Price lists, brochures and press releases often end with a named contact.
- Places the founder posts. For startups: Y Combinator company pages, "Who is hiring" threads on Hacker News, GitHub commit emails, conference speaker pages.
- A web search for
"@domain.com"in quotes.
When I looked at 80 hiring startups, 43 founders had published their address somewhere. About 10 of those were on the company site and 13 on YC pages or Hacker News. So the company site is a good start but not the only place.
Ignore addresses that clearly aren't staff: info@, hello@, support@, careers@. They tell you the domain works, not the person format.
Step 3: Apply the pattern
If you found anna@ for Anna Schmidt, Jonas Weber is very likely jonas@. If you found anna.schmidt@, he's jonas.weber@.
If you found nothing, here's what the data says about which guess to make. Of those 43 startup founders with a published address, 38 used plain firstname@. One used first.last@. That matches what I see on owner-run small businesses generally: the founder set up the domain, chose their first name, and never changed it.
Larger companies are different. Once there are two Davids, IT picks first.last@ or flast@ and applies it to everyone. So:
- Under ~20 people, owner-run: guess
firstname@. - Bigger, with an IT department: guess
first.last@unless you find evidence otherwise.
When I checked my tool's founder email guesses against 189 real people, 70% were exactly right. When the site showed its email format somewhere, that went up to 76%. Good enough to try; not good enough to treat as certain.
Normalise names before guessing: lowercase, strip accents (José → jose), and for German names use ue, oe, ae for umlauts (Jürgen → juergen). Drop middle names.
Step 4: Watch for catch-all domains
Some mail servers accept mail for any address on the domain, real or not. That's a catch-all. On a catch-all domain your guess won't bounce even if it's wrong, so "no bounce" proves nothing.
The common way to spot one is to ask the mail server, over SMTP, whether it would accept a made-up address like zq7x-not-real@domain. If it says yes, it's catch-all. The catch: many home and cloud networks block outgoing port 25, Microsoft and Google often answer "yes" to everything at that stage anyway, and repeatedly probing mail servers can get your IP flagged. I don't do this from my own machine. Paid verifiers do it from servers set up for it, and even they mark many results as "unknown".
What I do instead: if a guess is unverifiable, I label it as a guess in my sheet and keep my send volume low so a few bounces don't hurt my domain.
Step 5: Know when to give up
Stop guessing and use the shared inbox when:
- The domain has no MX records.
- The site lists only a contact form and you can't find one personal address anywhere.
- The company is clearly large and you have no name, only a role.
- The owner's name is common in two forms (Alex/Alexander, Kate/Katherine) and you have no example to choose between them.
An email to info@ that opens with "Could you pass this to Jonas?" gets read at many small businesses. A wrong guess to alexander@ gets nothing.
A quick word on rules: whatever address you end up with, the usual cold-email rules apply. In the US, CAN-SPAM wants a real sender and a way to opt out. In the EU and UK, GDPR and PECR apply to personal addresses like firstname@. Keep it relevant, send a few a day, and stop when someone asks.
Doing it for a list
The steps above are fine for ten companies. For a few hundred I use a tool I built, Founder Email Finder (https://founders.intropi.in, on Apify). It reads each company's site, takes a published address when there is one (labelled high confidence), and otherwise makes a firstname@ guess only on domains with MX records (labelled low). Pricing is $1.50 per 1,000 websites with contacts.
Disclosure: I built it. The method is the same one described here.
Written with AI assistance; the numbers come from my own test runs.