What Is an Email Pattern?

An email pattern is the naming rule a company uses to build employee addresses from names, such as first.last@acme.com. If you know the rule and a person's name, you can guess their address.

A guess is all it is. This page defines the term, walks through a worked example, shows where patterns fail, and explains how LeadOcean labels addresses built from one.

Key takeaways

• A pattern is a rule: name parts plus a separator plus the company domain.
• One company can run several patterns at once, so a rule learned from one address can be wrong for the next person.
• A pattern-built address is a candidate. Only a check against the mail server or a stored record makes it a fact.
• On LeadOcean, an address built from a pattern and never tested carries the status derived.

What it is

An email pattern is the rule a company uses to build the local part of staff addresses (the text before the @) from a person's name.

The rest of the address is the company domain. So jane.doe@acme.com follows the pattern "first name, dot, last name, at the domain". The pattern is first.last.

Nobody publishes these rules. IT admins pick one when they set up mail and the habit sticks. The mail standard says only the host named after the @ gives the local part its meaning. No outsider can read the rule from one address (RFC 5321, September 2026).

Common shapes, shown for Jane Doe at acme.com:

You will also hear "email format" and "company email format". They mean the same thing.

One more use of the word needs a flag. Some deliverability articles say "email patterns" when they mean how a mail server answers a probe, such as catch-all or greylisting. That is a different topic. We cover it in email pattern detection strategies for verification.

How it works

A tool or a person infers the rule from addresses that are already known, then applies it to a new name. The steps are the same whether a script or a sales rep does them.

1. Collect known addresses at one domain, from a database, a press page or a signature.
2. Split each local part and match it to the owner's name.
3. Find the rule that explains the most of them.
4. Apply the rule to the new person's name and domain.
5. Test the candidate before you send. Skip this step and you are mailing a guess.

Worked example, with placeholders. Three known addresses at acme.com:

john.smith@acme.com
maria.lopez@acme.com
wei.chen@acme.com

All three read first name, dot, last name. The rule is first.last. The new person is Jane Doe, so the candidate is jane.doe@acme.com.

Two questions are still open. Does that mailbox exist? And is it Jane, or a second Jane Doe at the same company? The pattern cannot answer either one.

Email pattern vs verified email

People treat a matching pattern as a working address. They are different kinds of evidence, and the gap is where bounces come from.

The catch-all row matters most. A catch-all server answers yes to any name, so a probe passes for a made-up pattern address too. Our guide on what a catch-all email is covers it.

The practical rule: a pattern tells you where to look. It does not tell you what is there.

When it matters

Cold outbound with no direct data

You have a name and a company and nothing else. A pattern gives you a candidate to test. Treat it as the last source you try, after a database lookup, and send only what survives a check.

Companies with mixed formats

After an acquisition, a company often runs two formats side by side. Staff from the acquired firm keep their old rule, and the rule you learned from five addresses fits only some of the people you want. Sample size is the control here. Three addresses can mislead you. A rule that explains most of a larger set is safer, and you still test the candidate.

Common names and collisions

If two people at one company are both John Smith, the second one cannot have john.smith@. IT adds a digit or a middle initial instead, and the pattern gives you the wrong address for them. Expect this at larger companies.

Executives and special accounts

Founders, long-tenured staff and shared mailboxes often sit outside the company rule. A CEO may have first@ while everyone else has first.last@. Do not trust a pattern for the people you most want to reach.

Personalization and list quality

Your merge fields assume the name is right. A pattern-built address can reach the wrong person with the right name in the greeting. Read the guides on email personalization and personalizing emails for how to keep that risk low.

How LeadOcean handles it

LeadOcean returns addresses from stored records, not from a pattern guesser. There is no name plus domain lookup: you search by domain first, then enrich by person_id (LeadOcean OpenAPI, September 2026).

Some records do carry an address built from the company's pattern. LeadOcean labels those derived, which the API defines as built from the company's address pattern and never tested (LeadOcean API enums, September 2026). The docs list it under "use with care", next to risky and role.

The field is email_status. The filter is emailStatus. Every people search row carries it as a flag, and sizing a search with count=true and limit=1 is free.

Two things keep derived addresses out of your send list:

• The default count in the MCP server covers verified, catch_all_valid and catch_all only. A derived address is outside it.
• GET /v2/people/email/work returns only addresses whose deliverability was checked: verified, catch_all_valid or catch_all. It never returns an untested address.

The call below sizes people with a checked address and spends no records.

curl -X POST "https://api.leadocean.io/v1/people/search?count=true&limit=1" \
  -H "x-api-key: $LEADOCEAN_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"emailStatus": ["verified", "catch_all_valid", "catch_all"], "domain": ["acme.com"]}'

To see the pattern-built group instead, send "emailStatus": ["derived"]. Passing emailStatus replaces the default, so name every status you want. The same filter works through leadocean_count_leads in the MCP server and on the Exports page of the app at app.leadocean.io, which shows the record price before you start.

Pricing is two plans: Free (1,000 records, one-off, no card) and Pro at $499 a month. See pricing.

FAQ