How to Write Boolean Search Strings for Recruiting (With Real Examples)

A boolean search string is a search query built from operators — AND, OR, NOT, quotation marks, and parentheses — that narrows a search engine, LinkedIn, or job board down to candidates matching very specific criteria at once, instead of scrolling through hundreds of loosely-related profiles. Most recruiters know the operators exist but never learn to combine them well enough to get precise results. Here's the actual syntax, with real examples you can copy and adapt.

The four operators that do almost all the work

AND narrows results — both terms must appear ("recruiter" AND "healthcare"). OR broadens results — either term is acceptable, and it's essential for synonyms ("recruiter" OR "talent acquisition" OR "TA"). NOT excludes a term ("developer" NOT "intern"). Quotation marks force an exact phrase match ("machine learning engineer" is different from machine AND learning AND engineer, which would also match someone with those three words scattered anywhere). Parentheses group logic together so OR doesn't silently break your AND — this is the single most common mistake in recruiter-written strings.

The parentheses mistake that breaks almost every beginner string

Without grouping, `nurse AND ICU OR "critical care"` doesn't mean what it looks like — it's read as `(nurse AND ICU) OR "critical care"`, which will return every critical-care hit regardless of whether the word "nurse" appears at all. The fix is explicit grouping: `nurse AND (ICU OR "critical care")`. Any time your string mixes AND and OR, wrap the OR group in parentheses — this one habit fixes most broken boolean strings.

A real worked example: sourcing a senior backend engineer

Goal: someone with backend experience in Python or Go, senior-level, not a contractor, not currently a manager. A working string: `("backend engineer" OR "software engineer, backend") AND (Python OR Golang OR "Go") AND (senior OR "staff engineer" OR "lead engineer") NOT (manager OR contractor OR intern)`. Notice the quoted exact phrases for multi-word titles, the OR groups in parentheses for synonyms and seniority variants, and the NOT clause doing real exclusion work at the end rather than being an afterthought.

A real worked example: sourcing on LinkedIn specifically

LinkedIn's search (both the free version and Recruiter) supports boolean syntax in the keyword field, but title/company/location fields are usually better filtered using LinkedIn's own structured filters rather than boolean — reserve boolean for the free-text keyword field where structured filters don't exist. A workable LinkedIn keyword-field string for a marketing candidate with SaaS experience: `("product marketing" OR "growth marketing") AND (SaaS OR "B2B software") NOT ("marketing intern" OR freelance)`. Combine this with LinkedIn's own seniority-level and location filters instead of trying to cram those into the boolean string too.

X-ray search: finding candidates Google-indexed on other sites

"X-raying" uses Google's `site:` operator to search a specific domain as if it were your own database — useful for finding public profiles on sites without good internal search. Example: `site:linkedin.com/in "data analyst" AND (Tableau OR "Power BI") AND "Chicago"` searches only LinkedIn public profile pages for that combination. This works on GitHub (`site:github.com`), personal portfolio sites, and conference speaker pages just as well — anywhere a candidate's public-facing profile exists and is indexed.

Common failure modes and how to fix each one

Zero results: your AND chain is too long or too specific — drop the least important term first, usually a nice-to-have skill rather than a must-have one. Too many irrelevant results: you're missing NOT clauses for obvious false positives (recruiter searches for "developer" without excluding "business development" get this constantly) or you forgot quotation marks around a multi-word phrase that's being matched as separate scattered words. Results that don't match your intent at all: check your parentheses grouping first — it's the most common silent failure, because the search still runs and returns results, just not the ones your logic intended.

Build a reusable template instead of starting from scratch each time

Most recruiters rewrite boolean strings from zero for every search. A faster approach: keep a personal template per role family — `(TITLE VARIANTS) AND (CORE SKILL VARIANTS) AND (SENIORITY) NOT (EXCLUSIONS)` — and swap in the specific terms per req. This is exactly the structure Rezumz's Boolean Search tool generates automatically from a role and job description, if you'd rather not hand-build strings for every requisition.