Does an ATS Recognize Synonyms? Exact Match vs Skills AI
If a posting asks for "client relationship management" and your resume says "account management," will the ATS connect them? The honest answer is: it depends which part of the system is doing the matching. Modern applicant tracking systems contain two very different kinds of matching, and their own documentation says one of them is literal and the other is not. Your resume has to work for both.
Layer one: keyword search is literal
When a recruiter types a search, most documented ATS search features look for the string. Greenhouse's Talent Filtering says so directly: "To appear in your results, the keyword from your search must exactly match the keyword in the application."
Synonyms enter only when the recruiter adds them. Greenhouse's Boolean guide illustrates OR with "trainer OR instructor OR teacher," which "will return results that contain one or more of the stated words that are synonymous." The system is not inferring that these words mean the same thing; the recruiter is listing them. Word forms are handled the same way. Greenhouse supports a wildcard, so Java* returns Java and Javascript, but only if the recruiter types the asterisk.
LinkedIn Recruiter is stricter still. Its Boolean guide says wildcards such as an asterisk are not supported, and its filter help lists stop words that the Keywords field ignores. A recruiter who wants "analyst" and "analytics" has to type both.
The practical meaning: if a recruiter searches for the posting's phrase and your resume uses a different phrase, you will not appear in that search, however equivalent the two are.
Layer two: skills AI maps related terms
The newer matching features are built precisely to get past that literalism.
Greenhouse's Talent Matching FAQ explains that for skills, "it compares related terms on each resume to the calibrated skills you set, and multiple terms can map to the same calibrated skill." On the recruiter's screen, the resume preview highlights "exact keyword matches and similar keyword matches" separately (Talent Matching). The same FAQ notes the system "warns the user if a skill can't be parsed from the candidate's resume," which tells you the mapping has limits.
Workday's approach is an ontology. Its Skills Cloud datasheet describes "a repository of more than 73,000 standardized skills" and the ability to "infer relationships between different skills." Workday's Automated Skill Management documentation describes background jobs that find skills which are synonyms of a Skills Cloud skill and migrate them to the standard one, though it says skills used by candidates are not changed. Its recruiting setup guide adds that Workday keeps the skill string a candidate enters, and that recruiters can filter candidates by Skills Cloud skills.
LinkedIn's Job Match, launched in January 2025, is described the same way. Engadget reported that, according to LinkedIn product manager Rohan Rajiv, it "goes beyond the kind of simple keyword matching that job hunters may already rely on."
| Matching layer | Example feature | Recognises synonyms? |
|---|---|---|
| Recruiter keyword filter | Greenhouse Talent Filtering | No: exact match required |
| Recruiter Boolean search | Greenhouse full-text, LinkedIn Recruiter | Only the ones the recruiter types with OR |
| Wildcard search | Greenhouse Java* | Word forms only, if the recruiter uses *; not in LinkedIn |
| AI skill matching | Greenhouse Talent Matching | Yes: related terms map to one calibrated skill |
| Skills ontology | Workday Skills Cloud | Yes: synonyms consolidated to standard skills |
| Job seeker match | LinkedIn Job Match | Described as going beyond simple keyword matching |
Why you cannot rely on the smart layer alone
Both layers run in the same employer's system, and you do not know which one a recruiter will use on your application. A recruiter might open the AI-ranked list, or might type the one tool that matters into a filter. Greenhouse lets recruiters do either, and its Talent Filtering even offers "suggested keywords generated from your public job post," which means the literal search terms frequently come from the posting's own wording.
The large-employer picture makes this concrete. Greenhouse and Workday are both common among big US employers: in the ATS Resume AI analysis of all 503 S&P 500 careers sites and 9,260 live job postings, June 2026, Workday ran at 87 companies and Greenhouse at 23 (full breakdown). Both vendors document literal and AI matching side by side.
Acronyms, spellings and job titles
Three kinds of variation cause most misses in literal search.
Acronyms. "SEO" and "search engine optimization," "AWS" and "Amazon Web Services." Write the full form once with the acronym in brackets, then use whichever reads better.
Spellings and spacing. "E-commerce" and "ecommerce," "JavaScript" and "Java Script." Use the posting's form.
Job titles. "Client Success Manager" and "Customer Success Manager" may be the same job, but a quoted title search matches only one. Our guide to changing a job title for ATS covers when using the posting's title is honest.
A worked example
Customer-facing roles are full of near-synonyms: customer success, client success, account management, customer support. In the ATS Resume AI analysis of all 503 S&P 500 careers sites and 9,260 live job postings, June 2026, "customer success" appeared in 76.5% of full-text customer service postings and "account management" in 35.3% (customer service data). A resume written in a company's internal vocabulary can easily use neither.
Before:
Client Partner, Northwind · Looked after a book of 40 client relationships; ran onboarding and renewals.
After:
Client Partner (Customer Success), Northwind · Managed customer success and account management for 40 client relationships; ran onboarding and renewals.
The job and the facts are unchanged. The posting's two phrases now appear once each, so an exact-match filter for either one finds the resume. "Client Partner" stays, because it is the real title, and AI skill matching can still relate "client relationships" to the calibrated skill. Nothing was added that the person did not do. That is the whole technique: translate your vocabulary into the employer's once, and leave the rest of the resume in plain, accurate language.
Cover both layers in five steps
- List the posting's key terms exactly as written, including its acronyms and its spelling of each tool.
- Compare each to your resume with Find. Mark exact matches, near matches (a synonym or different form) and misses.
- For each near match, add the posting's form once. Put it in your skills line or in brackets after your own term: "client relationship management (account management)."
- Leave your natural wording in the bullets. AI matching maps related terms, and a human reads better prose than a list of the employer's phrases.
- Do not add terms for skills you lack. A synonym is a different word for something you did, not a way to claim something you did not.
The free resume keyword generator pulls a posting's key terms if you want a starting list, and our ATS resume keywords guide covers placement.
Sources
- Greenhouse Support, "Talent Filtering"
- Greenhouse Support, "Search candidates using Boolean queries"
- Greenhouse Support, "Talent Matching"
- Greenhouse Support, "Talent Matching FAQ"
- LinkedIn Help, "Use Boolean search on LinkedIn"
- LinkedIn Help, "Use Boolean to filter search results in Recruiter and Recruiter Lite"
- Workday, Skills Cloud datasheet
- Workday, "Concept: Automated Skill Management"
- Workday, "Setup Considerations: Skills in Recruiting"
- Engadget, "LinkedIn wants you to apply for fewer jobs" (2025)
Frequently asked questions
- Does an ATS understand that 'managed' and 'management' are related?
- Only in some layers. Greenhouse full-text search treats them as different words unless the recruiter uses a wildcard such as manag*, and LinkedIn Recruiter does not support wildcards. AI skill matching is designed to relate such terms.
- Should I write both the acronym and the full term?
- Yes, once. Literal search matches whichever form the recruiter types, so writing 'Search Engine Optimization (SEO)' a single time covers both without repetition.
- Is semantic AI matching replacing keyword search?
- Not in the documentation. Greenhouse offers exact-match Talent Filtering and AI Talent Matching side by side, and Workday offers skills filters alongside its ontology. Recruiters choose which to use.
- Will the ATS change the skills I typed into a standard name?
- Workday's documentation says it keeps the skill string a candidate enters, and its synonym clean-up does not change skills used by candidates. Recruiters may still search by the standard Skills Cloud name, so use common skill names.