TL;DR
- Two company records with the same clean domain are often duplicates.
- The domain after the @ in a work email points to the contact's company.
- Compare websites on their core domain, so formatting differences don't hide a match.
- Personal email domains like gmail.com can't link a contact to a company.
- Dedupely finds duplicate companies and contacts by matching the core domain of emails and websites.
Company names are messy. One rep types "Acme," another types "Acme Inc.," and an enrichment tool adds "ACME Corporation."
The domain is usually the steadiest piece of data a company record has, and every work email carries one. That makes it a good place to start when you look for duplicate companies, and for contacts linked to the wrong one.
It also has limits: personal emails, shared domains, and companies with a domain for each country. In this post, you'll see how to get a clean domain and match on it. You'll also see where domain duplicates come from and which domains to leave out.
Why the domain is a strong link between records
A domain is registered once, so two companies rarely share it by accident. Names, on the other hand, are typed by people and tools in many different ways, often with company suffixes like Inc. and LLC.
What a domain tells you that a name doesn't
"Northwind," "Northwind Traders" and "Northwind Traders LLC" may or may not be one company. If all three records list northwindtraders.com, you have strong evidence they are.
The domain also connects objects. A contact's email domain can point to the company record it belongs to, even when nobody filled in the company field.
Getting a clean domain out of an email or URL
From an email, the domain is everything after the @. From a website field, you need to strip the parts that vary from record to record:
- The protocol: https:// or http://
- The www. prefix
- Paths, pages and tracking parameters after the domain, like /contact or ?utm_source
- Capital letters and trailing slashes
Once you do that, "https://www.Northwind.com/about" and "northwind.com" become the same value, and an exact match catches them.
Start with the obvious matches
Before anything else, look for records with the exact same clean domain. These are usually your most obvious duplicates and the safest to merge first.
Looser matches, like country domains or subdomains, come after you've seen what the exact matches look like in your own data.
Following contacts to duplicate companies
Contacts carry a domain in every work email, even when the company field is empty. That makes them a useful trail to duplicate companies.
Several contacts, one domain, different companies
When ten contacts share a domain, they usually work at one company. If those ten contacts point to three different company records, those company records deserve a closer look.
Group your contacts by email domain and see how many company records each group is linked to. A domain linked to several companies often hides a duplicate company.
Duplicate contacts at one domain
The domain helps with contacts too, as long as it isn't the only field. Domain alone groups colleagues. Paired with the name, it finds one person with two work emails, like jane@northwind.com and j.doe@northwind.com.
When CRMs link contacts to companies for you
Some CRMs use the domain to link contacts to companies automatically. HubSpot, for example, can match a contact's email domain to a company's domain and associate the two. If no company with that domain exists, it can create one.
As of October 2026, from HubSpot's knowledge base: Automatically create and associate companies with contacts.
Where domain duplicates come from
Two company records with the same clean domain are a common company duplicate. How they got there varies from CRM to CRM and from team to team.
Enrichment tools
Enrichment tools are a frequent source. A tool can create a new company from a contact's email domain without finding the record that already exists. Often the existing one stores the website in another format.
Automatic company creation
In HubSpot, each unique domain gets its own company record, and subdomains count as separate domains. So @example.com and @info.example.com can produce two companies.
As of October 2026, from HubSpot's knowledge base: Automatically create and associate companies with contacts.
Forms that name a different company
From what I've seen, a signup can also leave a contact linked to two companies. It happens when the email domain and the company typed in the form don't agree.
Following the domain shows which link came from the email and which from the form. Then you can decide which company the contact belongs to, and whether either company record is a duplicate.
Same company, different-looking domains
Country domains like northwind.com and northwind.de, and subdomains like info.northwind.com, can hide one company. Whether they count as one depends on your team. A company that sells to its German office separately may keep northwind.de apart.
When the same domain isn't a duplicate
Subsidiaries, regional branches and franchises often share a parent's domain. What one team considers a duplicate, another may not, so add a second field like city or country when branches should stay apart.
Personal email domains don't work for domain matching
A contact at gmail.com, outlook.com or yahoo.com tells you nothing about their company. If you match on those domains, thousands of unrelated contacts end up in one giant match.
Filter generic domains out
Before matching on domain, filter out generic email providers so they never come up as matches. The common ones are easy to list: gmail.com, outlook.com, hotmail.com, yahoo.com, icloud.com and a few regional providers.
Also check for domains specific to your market, like a local internet provider many of your customers use.
What to do with personal-email contacts
For contacts with a personal email, match on other fields. Name plus phone, or name plus company name, can still find duplicates among them.
Some CRMs fall back on another field here. HubSpot, for example, checks the contact's website URL when the email uses a freemail domain.
As of October 2026, from HubSpot's knowledge base: Automatically create and associate companies with contacts.
Domain matching at a glance
| Situation | Does the domain help? | What to watch |
|---|---|---|
| Same clean domain on two company records | Yes, often an obvious duplicate | Subsidiaries and branches that share it |
| Contacts on one domain linked to several companies | Yes, points to duplicate companies | Contacts linked to two companies by a form |
| Two contacts with the same name on one domain | Yes, with the name as a second field | Colleagues who share a name |
| Country domains or subdomains | Sometimes | Teams that sell to each country separately |
| Personal email domains | No | Filter them out and match on other fields |
How Dedupely finds duplicates by domain
In Dedupely, you build duplicate searches that match the core domain of an email or website, so northwind.com, www.northwind.de and sales@northwind.com count as one. Add a second field like country to keep branches apart, and filter out generic domains like gmail.com. When two records share a domain but aren't duplicates, reject the match and it won't come back in your searches.
Let the domain do the first round of matching
The domain is the closest thing many CRMs have to a shared key between contacts and companies. A clean domain, compared exactly, finds your most obvious company duplicates and the contacts that point to them.
From there, the decisions are yours: which country domains belong together, which branches stay apart, and which generic domains to leave out. Once you know what counts as a duplicate in your own CRM, the domain becomes a reliable first step.
Frequently asked questions
How do I find duplicate companies by domain?
Clean the website field on every company record by removing https, www, paths and capital letters. Then compare the clean domains: records with the same one are likely duplicates. Check for subsidiaries or branches that share a domain before merging, since some teams keep those as separate records.
How do I extract the domain from an email address?
Take everything after the @ sign and lowercase it: for Jane@Northwind.com, the domain is northwind.com. Lowercasing keeps capital letters from blocking a match. Generic providers like gmail.com or outlook.com won't link a contact to a company, so use other fields for those contacts.
Why is a contact associated with two companies?
Often because two sources disagree. A form may link the contact to the company typed in, while automatic association links it to the company behind the email domain. Compare the email domain with each company's domain to see which link is right, then remove the other.
Why shouldn't I match contacts on gmail.com or other personal domains?
Personal email domains are shared by millions of unrelated people, so matching on them groups contacts who have nothing in common. Filter generic domains out before matching on domain. For contacts with personal emails, find duplicates with other fields like name and phone.
Should country domains like .com and .de count as one company?
It depends on your team. If you manage a company as one global account, northwind.com and northwind.de are likely one record. If each country office has its own rep, deals and contracts, keep them apart, and decide this before matching on the core domain.
Or sign up with Google
Salesforce AppExchange 
