Ignored Terms
Ignored terms are words that are skipped when two values are compared during duplicate matching. They're words that show up across many records without saying which person or company a record describes: legal suffixes such as Inc, LLC, Ltd or Sp. z o.o., generic words such as "Group" or "The," and titles such as Mr. or Dr. Skipping them lets values that differ only in those words match.
Ignored terms are also called stop words or noise words, and skipping them is a common part of company name normalization. They make a comparison more lenient, so what goes on the list matters. A word that looks generic can be the only thing that tells two records apart, and ignoring it pairs records that aren't duplicates. That's why ignored-term lists work best when they're short and checked against the values a field really holds.
See how Dedupely skips legal suffixes and generic words in your CRM
What Are Ignored Terms?
Ignored terms are the words a matching rule leaves out of a comparison because they don't identify anything. Take two company records: "Acme Logistics Inc" and "The Acme Logistics Group." Compared as whole values, they're different. With "Inc," "Ltd," "The" and "Group" ignored, both read "Acme Logistics" and match. Add "Logistics" and "Holdings" to the list as well, and "Acme Holdings Ltd" reads as "Acme" too, so a different company now pairs with both.
Company names are the best-known use, but not the only one. Person names carry titles and honorifics, deal names carry words like "Renewal" or "Deal," and addresses carry words like "Street" or "Suite." The idea applies to every object a CRM stores: companies in HubSpot, accounts in Salesforce, organizations in Pipedrive, leads in Close and contacts in Nimble or Google Contacts.
There are a few common approaches. A built-in list ships with the tool and covers common suffixes and stop words. Its strength is that it works without setup. A custom list is built by the team for each field. Its strength is that it fits the data's own words, such as regional suffixes like GmbH or industry terms, and skips only what the team chose. Frequency weighting skips nothing and gives words that appear in many values less weight in a match score. Its strength is that there's no list to keep up, and a common word still counts a little. Whichever approach is used, the goal is the same: compare the words that identify a record, and leave out the ones that don't.
How Ignored Terms Work
Ignored terms change what gets compared, not how it's compared. They work in a few steps:
- Build the list for a field. The list holds words that repeat across many values without identifying anyone: legal suffixes, articles such as "The," generic words such as "Group" or "Company," and titles. Each field needs its own list, because the words that don't matter in a company name differ from the ones in a person's name or a deal name.
- Cover the variants. The same suffix is often written several ways, such as "Inc," "Inc." and "Incorporated." Depending on how a tool treats punctuation and abbreviations, each form may need its own entry.
- Skip the terms in both values. Before the comparison, the listed words are left out of each value. This works best on whole words, so ignoring "Inc" leaves a name like "Incline Partners" untouched.
- Compare what's left. The remaining words are compared with the field's method, whether that's exact, similar, phonetic, word-based or domain matching.
- Handle values with nothing left. A value made up only of ignored terms, such as "The Group," has nothing to compare. Tools handle this differently, so it's worth checking how short values behave.
Every term added makes the field more lenient. That's useful while the words are true noise, and risky once they aren't. Over-ignoring means skipping a word that tells two records apart: "Jr." and "Sr." on a father and son, "North" and "South" on two branches, or "Inc" and "GmbH" on two entities a company tracks separately. Those records then match as duplicates.
Why Ignored Terms Matter for CRM Teams
Ignored terms decide whether a whole class of duplicates gets found. Company names reach a CRM from web forms, imports, integrations and reps typing them in, and each source writes legal suffixes and generic words its own way. Without ignored terms, "Acme Inc" and "Acme Incorporated" stay two records. Account counts are inflated, contacts and deals are split between the two, and reports show two customers where there's one. In CRMs that bill by record count, each one is also a record paid for twice.
Over-ignoring causes the opposite problem, and it's harder to see. Each word added to the list pairs more records, and some of those pairs aren't the same person or company. A subsidiary and its parent, two regional branches, or a father and son with the same name can end up in one duplicate set. If that set is merged, their owners, deals, activity and contact details are mixed into one record, and in many systems a merge is hard or impossible to reverse.
A short, checked list keeps both risks low. It catches the duplicates that differ only in noise words, and leaves alone the words that carry meaning in a given CRM. Because those words differ from one database to the next, a list that works for one team can create false matches for another.
Getting Started With Ignored Terms
Start from the data, not from a generic list. Look through the values in the field you want to match and note the words that repeat across many records without identifying anyone. On company names, that's often the legal suffixes in the forms your data really uses, plus articles and words like "Group." On person names, it's often titles.
Before adding a word, ask whether it ever tells two records apart in your CRM. If you track subsidiaries or regional entities as separate accounts, a legal suffix or a word like "North" may be the only difference between them, so leave it off. Back the loosened field with a stronger one, such as website domain, email or phone, so a match on the name alone doesn't pair two different companies.
Add terms a few at a time and compare the results with and without them. The new duplicate sets are the ones to check, because that's where false matches show up. Reject the pairs that aren't duplicates so they don't come back, and take a term off the list if it keeps producing them.
In Dedupely, Ignored Terms are an optional setting on each field of a Duplicate Search, with up to 50 terms per field, and they work with every Match Option except None. That lets you skip legal suffixes on a company name compared with Similar Words, for example, while Domain Root on the website field keeps the match tied to one company.
Key Takeaways: Ignored Terms
Ignored terms are words skipped when values are compared, such as legal suffixes, generic words like "The" or "Group," and titles. Skipping them lets records match when their values differ only in those words, which matters most on company names but also helps on person names, deal names and addresses. Each term added makes a field more lenient, and ignoring a word that tells records apart creates false matches, so short, checked lists work best. In Dedupely, Ignored Terms are optional, with up to 50 per field, on every Match Option except None. Similar Words and Domain Root suit company names, Rejected Matches keep false pairs out of future results, and Safety Settings hold back matches with more than 20 records.
Frequently Asked Questions About Ignored Terms
Which words should you add to ignored terms when matching company names?
Add the words that appear in many company names without identifying the company: legal suffixes, articles and generic words. Legal suffixes include Inc, LLC, Ltd, Corp, GmbH and Sp. z o.o., in each form your data uses, such as "Inc," "Inc." and "Incorporated." Articles like "The" and generic words like "Group," "Company" or "Holdings" belong on the list when they don't carry meaning in your CRM. Leave off any word that tells two of your accounts apart, such as a suffix that separates a parent from its subsidiary or a word like "International" that marks a separate entity. Industry words like "Logistics" or "Consulting" are risky too, because they can be most of what's left of a short name. In Dedupely, the Similar Words Match Option compares shared words in Any Order, or the First or Last words, and you can add up to 50 Ignored Terms to the field. Domain Root on the website field then confirms the company behind the match.
When should you use ignored terms on fields other than company names?
Use ignored terms on any text field where some words repeat across records without identifying them. Person names often carry titles like Mr., Ms. or Dr., so "Dr. Jane Smith" and "Jane Smith" stay apart until the title is ignored. Deal names often carry words like "Renewal," "Upsell" or the year, address fields carry words like "Street," "Suite" or "Floor," and job titles carry words like "Senior." The same caution applies on every field: "Jr." and "Sr." look like noise but tell a father and son apart, so they're better left off a name list. It helps to back the loosened field with a stronger one, such as email or phone. In Dedupely, Ignored Terms are available on every Match Option except None: Exact, Similar, Fuzzy, Similar Words and Domain Root. That means you can skip titles on a name field compared with Fuzzy, or legal suffixes on a company name compared with Similar Words.
How do you ignore Inc, LLC and Ltd when matching company names in a CRM?
Add the suffixes to an ignored-terms list on the company name field, so they're skipped before the names are compared. With "Inc," "LLC" and "Ltd" ignored, "Acme Inc" and "Acme" match, whether the records are accounts in Salesforce, companies in HubSpot or organizations in Pipedrive. Cover each form the suffix takes in your data, because "Inc" and "Incorporated" are different words, and depending on the tool, punctuation may make "Inc." a different entry too. Pair the name with a word-based method, so names that also differ in word order still match, and with a second field such as website domain or phone. If your CRM keeps entities that differ only by suffix as separate accounts, ignoring suffixes will pair them, so check those sets before merging. In Dedupely, you add Inc, LLC and Ltd as Ignored Terms on the company name field, up to 50 per field. Similar Words then compares the remaining words in Any Order, or the First or Last words, with a Number of Words to Match setting.
Why does ignoring too many words cause false matches?
Because every ignored word removes something that could tell two records apart, and with enough of them gone, different records look the same. Over-ignoring often starts with words that seem generic but carry meaning: "Holdings" and "Logistics" on two companies that share a brand name, "North" and "South" on two branches, or "Jr." and "Sr." on two people. Once those words are skipped, the values that are left can be short and common, so one match can pull many unrelated records into one large duplicate set. Merging such a set mixes their histories, owners and contact details. To avoid it, keep the list short, back the field with a stronger one and check the sets each new term creates. 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 values, in an Excluded tab. Rejected Matches keep false pairs out of current and future Duplicate Searches, and you can send uncertain sets to Manual Review.
Should you ignore terms when matching or clean company names in the CRM?
Ignoring terms when matching finds duplicates without changing the data, and cleaning names in the CRM standardizes them for everything else, so the two work well together, with matching first. Ignored terms change only what's compared, so the stored names stay as they are, and a list that's too loose can be changed without touching a record. Company name normalization in the CRM itself rewrites the stored values, for example by removing or standardizing suffixes, which helps reporting, routing and other tools that read the name. It changes the data, though, so it's worth doing once duplicates are merged and the team agrees on one format. In Dedupely, the Match Options guide notes that changing match options never changes CRM data until a merge starts. When a merge does run, Merge Rules decide which value each field keeps, including the company name.
