Domain Root Matching
Domain root matching is a record-matching method that compares the core domain of an email address or website instead of the full value. Before the comparison, it extracts the domain and sets aside the parts that don't identify the organization: the www, subdomains such as mail. or shop., and the ending, from .com to country variants like .mx or .jp. Two records match when what's left, the root, is the same.
Because one company can use many addresses and websites, the full values often differ even when the organization is the same: jane@acme.com and sales@eu.acme.co.uk are different values, but they share a root. That makes domain root matching a common way to link contacts to the company they work for, and to find company records entered twice with different websites. It works best alongside a second field, because the domain alone doesn't tell people apart.
See how Dedupely matches records by domain root in your CRM
What Is Domain Root Matching?
Domain root matching is the method that answers one question about two records: do they point to the same organization online? It goes by several names, including root domain matching, email domain matching and domain matching, and its first step, domain extraction, is often searched for on its own.
Take three records: a company saved as "Acme Inc., https://www.acme.com," another saved as "Acme México, acme.mx," and a contact with the email j.smith@mail.acme.co.uk. Compared as full values, none of them match. Once the domain is extracted and reduced to its root domain, all three come down to "acme," so the two companies are flagged as possible duplicates and the contact can be tied to them.
Tools take two common approaches to the root. Registrable-domain matching keeps the ending, so acme.com and acme.co.uk stay apart. Its strength is caution: a company in one country is never paired with an unrelated business that owns the same name in another. Core-name matching drops the ending as well, so acme.com, acme.co.uk and acme.mx all match. Its strength is reach: it finds regional sites and international branches of one company.
The method applies to the objects that hold websites and emails: companies in HubSpot, accounts in Salesforce, organizations in Pipedrive, leads in Close, and contacts in Nimble and Google Contacts.
How Domain Root Matching Works
Domain root matching starts with domain extraction from emails and URLs, which follows the same idea in any tool, and then compares what's left:
- Find the domain. In an email address, the domain is everything after the @. In a URL, it sits between the protocol (https://) and the first slash, so the path, page and tracking codes are dropped. When a field holds free text with an email or URL inside it, the address has to be found first.
- Even out the format. Capital letters, a trailing dot or a trailing slash are removed, so ACME.com/ and acme.com read the same.
- Remove the www and subdomains. Prefixes like www., mail., shop. or eu. describe a server or a section of a site, not a different organization.
- Handle the ending. Endings vary in length: .com has one part, .co.uk and .com.br have two. Keeping only the last two parts turns sales@acme.co.uk into "co.uk," so reliable extraction checks a list of known endings, including country ones. Depending on the approach, the ending is then kept or dropped.
- Compare the roots. Records with the same root are paired, often only when another field, such as company name, also matches.
A blank or invalid value has no root to extract, so it doesn't make a match. Shared domains need care too: everyone with a free email address shares that provider's root, and subdomains on a hosting platform reduce to the platform's root.
Why Domain Root Matching Matters for CRM Teams
A company's domain is one of the few values its people and its records have in common. When email domain matching ties contacts to the right company, activity, deals and owners roll up to one account, and reports on accounts, pipeline and revenue reflect the real customer instead of a scattered set of contacts. Contacts left unlinked fall outside account-level views, routing rules and territory assignments.
Company duplicates often hide behind different websites. One record holds www.acme.com, another acme.mx from a regional team, a third the domain from a signup form. Each one splits the company's history, can put two reps on the same customer, and in CRMs that bill by record count, adds a record that's paid for more than once. A comparison of the full website value doesn't pair them; a comparison of the root does.
The same breadth creates risk. A root shared by unrelated records, such as a free email provider or a hosting platform, can tie many different people or businesses into one set, and a merge built on it mixes their data. Core-name matching can also pair two unrelated companies that use the same name in different countries. That's why domain root matching is often paired with a second field and checked before records are merged.
Getting Started With Domain Root Matching
Start with the fields that hold domains: the website field on company records and the email field on contacts. Before you extract the domain from email addresses and websites, check how the values are stored. Some records hold a full URL with a path, some only a bare domain, and some hold a placeholder such as "N/A." Knowing the mix tells you how much the extraction has to handle.
Decide early whether country domains should match. If acme.mx and acme.jp belong to one customer in your CRM, match on the core name. If regional entities are kept as separate accounts, keep the ending or add a field such as country to the rule. Then pair the domain with a second field, such as company name, so a shared root alone doesn't pair records.
Before merging, look for roots that appear on many unrelated records, such as free email providers, shared mailboxes or hosting platforms. Leave those records out of the comparison or rely on other fields for them, and check a sample of the sets each search finds before acting on them.
In Dedupely, Domain Root is one of the Match Options in a Duplicate Search: it compares the core domain of an email or website and ignores www, .com and country variants. Pair it with a second field that has its own Match Option, such as company name on Similar Words: when a search compares both fields, both must meet their criteria before records match.
Key Takeaways: Domain Root Matching
Domain root matching compares records on the core domain of an email address or website. Domain extraction removes the www, subdomains and the ending, so values that look different, such as acme.com and sales@acme.mx, can match. It links contacts to their company and finds company duplicates entered with different websites. Approaches differ on country endings, and shared roots such as free email providers need a second field or a closer look. In Dedupely, Domain Root is one of the Match Options in a Duplicate Search, and it ignores www, .com and country variants. Safety Settings hold back matches with more than 20 records, Merge Rules decide what each merge keeps, and Auto Match and Auto Merge keep trusted searches running.
Frequently Asked Questions About Domain Root Matching
How do you match contacts to a company by email domain?
Extract the domain from each contact's email address, reduce it to its root, and compare it with the root of the company's website. A contact with jane@eu.acme.com and a company with www.acme.com share the root "acme," so the contact belongs with that company. In CRMs such as HubSpot or Pipedrive, contacts are linked to a company record, and some CRMs can create that link by email domain, but what each one does differs, so the CRM's own documentation is the place to confirm. Contacts with free or personal email addresses can't be placed this way, because their domain doesn't belong to their employer. In Dedupely, matching and merging happen within one object, so it doesn't link a contact to a company record. On contacts, Domain Root on the email field, combined with a name field, finds one person saved with two addresses at the same company. On companies, Domain Root on the website field finds company records that share a core domain.
Which domains should you leave out of domain root matching?
Leave out roots that many unrelated records share. Free email providers are the clearest case: thousands of different people can share one provider's domain, so a match on it says nothing about who they are. Shared mailboxes from a reseller or agency, subdomains on a hosting or store platform, and placeholder values such as "none.com" or "N/A" cause the same problem. To find them, list the most common roots in your data and look at any that cover many unrelated records. Then filter those records out of the search, or match them on another field, such as full email address or phone number. In Dedupely, Safety Settings are on by default and hold back matches with more than 20 records, as well as matches built on poor or placeholder data, in an Excluded tab. From there, you can merge them by hand, reject them or send them to Manual Review.
How does domain extraction turn an email address or URL into a root domain?
Domain extraction takes the domain out of the full value, then strips the parts that don't identify the organization. For an email, it keeps what follows the @. For a URL, it drops the https://, the path and anything after it. Then it removes the www and other subdomains, and evens out capital letters. The trickiest step is the ending, because endings come in different lengths: .com has one part and .co.uk has two, so a rule that keeps the last two parts gives the wrong root. Reliable extraction checks a list of known endings, including country ones, before deciding what the root is. For deduplication, extraction is only the first half: the roots still need to be compared across records, along with another field. In Dedupely, Domain Root handles extraction and comparison in one Match Option. You pick it for an email or website field in a Duplicate Search, and it compares the core domain while ignoring www, .com and country variants.
When should country domains like acme.mx and acme.jp count as the same company?
Treat them as one company when your CRM tracks the organization as a whole, and keep them apart when regional entities are separate accounts. A parent company with regional sites, such as acme.com, acme.mx and acme.jp, is one customer for many teams, so matching on the core name finds those records and brings their history together. If each country office has its own owner, contract or billing, merging those records loses that separation. Core-name matching can also pair unrelated businesses that happen to share a name in different countries, so a second field, such as company name or phone number, helps confirm the match. In Dedupely, Domain Root ignores country variants, so acme.mx and acme.jp match. To keep country domains apart, use Exact on the website field instead, which compares the full value while ignoring capitalization. Or add a country field with its own Match Option: when a search compares both fields, both must meet their criteria.
Why does domain root matching miss company duplicates?
Often, one of the records has no usable domain to compare. The website field may be blank, hold a placeholder, or contain a value that isn't a valid URL. A company may also use two unrelated domains after a rebrand or an acquisition, and a contact may sign up with a personal email address. In each case, the roots differ or there's nothing to extract, so the records don't pair. Typos, such as acme.com and acmee.com, also produce two different roots. To find the cause, compare the domain fields on the two records and check whether another field, such as company name or phone number, could carry the match instead. In Dedupely, "Send incorrect formats to manual review" is a per-field checkbox, off by default. When checked, invalid values and filler values such as blanks or N/A go to Manual Review automatically. Typo allowance, one of the search's Enhancements, keeps small differences from blocking a match and applies to any Match Option.
