Boolean Search for Recruiters: Operators, Strings and What Still Works in 2026

You've been handed the job order nobody else wants. You pasted in a Boolean string from a sourcing blog, hit enter, and got either nothing or 40,000 recruiters. Sound familiar?

Here's the uncomfortable part: a lot of the Boolean advice still circulating was written for a Google that no longer exists. Google quietly retired operators that recruiters were taught to rely on, and most guides never updated. Recruiters spend an average of 14.6 hours a week searching for the right candidates, according to Bullhorn's GRID 2025 research (more than 1,500 respondents, February 2025). That's a lot of hours to spend on strings that half-work.

This guide gives you the operator set Google actually documents, 12 ready-to-run strings you can copy in one click, and a straight answer to the question everyone's really asking: does any of this matter now that search is going semantic?

What Boolean Search in Recruitment Actually Is

The logic is named after George Boole, a 19th-century mathematician. That's all the history you need. In recruiting, Boolean search means telling a search box exactly which words a profile must contain, which ones are interchangeable, and which ones rule it out.

Four building blocks do almost all the work:

• AND — every term must appear. React AND TypeScript returns fewer, tighter results.

• OR — any of the terms can appear. Developer OR engineer catches both ways people describe the same job.

• NOT (or the minus sign) — removes anything containing the term. Useful for stripping out recruiters and job ads.

• Grouping — quotation marks lock a phrase together; parentheses make OR apply to the right words.

Even if Google changes the rules again, this logic isn't going anywhere. It's how most ATS keyword searches, job board résumé databases and LinkedIn's own search filters are built. Google is just the place where the rules moved most without anyone announcing it loudly.

The Google Boolean Search Operators That Still Work

Google's own Refine web searches help page documents six operators: quotation marks, site:, the minus sign, before:, after: and filetype: (it also shows before: and after: combined as a date range). OR, AND, +, ~, * and related: don't appear on it. That's the defensible claim: not "only these work," but "these are the ones Google publishes."

Here's the full cheat sheet, side by side with LinkedIn and a typical ATS. Bookmark it — or link to it.

Last verified: October 14, 2026. Google column checked against Google Search Help's Refine web searches page. LinkedIn behaviour differs by product — Recruiter, Recruiter Lite, jobs search and Talent Insights each have their own help page — so confirm against LinkedIn's help docs for the product you use. ATS support varies by vendor.

The operators Google retired

• + was removed in October 2011. Google's advice at the time was to use quotation marks instead, having "expanded the functionality of the quotation marks operator" (Search Engine Land, 2011).
• ~ (synonyms) was dropped in mid-2013. Google's Dan Russell explained it was "because too few people were using it to make it worth the time, money, and energy to maintain" (Search Engine Land, 2013).
• AND, OR and * aren't on Google's documented list. Behaviour varies, and you shouldn't build a critical search on them alone.

If your saved strings are full of + and ~, they're not doing what you think. They're doing nothing, quietly, which is worse.

Why site: doesn't return everything

That one sentence explains most X-Ray frustration. LinkedIn is about as big as sites get. Your site: search was never going to show you every matching profile — it shows you a slice. The fix isn't a cleverer string. It's running several narrower strings and comparing what comes back.

LinkedIn Boolean Search vs Boolean on Google

LinkedIn is still where most sourcing happens: in the 2024 Employ Recruiter Nation Report (1,200+ talent acquisition decision makers), 71% of recruiters named LinkedIn among their most valuable social channels. But "Boolean on LinkedIn" means two different things.

LinkedIn's own search

Inside LinkedIn, you're searching LinkedIn's database with LinkedIn's rules. Sourcers typically type uppercase AND, OR and NOT, quotation marks and parentheses in keyword fields. The catch: operator support differs by LinkedIn product. Recruiter, Recruiter Lite, jobs search and Talent Insights each have their own official help page on Boolean, and they don't all say the same thing.

We're not going to paraphrase those pages from memory, because this is exactly the kind of detail that changes. Open LinkedIn's help page for the product you're paying for and check it before you build a search you'll reuse.

X-Ray search: searching LinkedIn through Google

X-Ray search uses Google's site: operator to search inside another site — most often site:linkedin.com/in, which is the path for public profiles. You're searching Google's copy of those pages, not LinkedIn itself, so you can reach people outside your network using Google's operator rules instead of LinkedIn's.

Why X-Ray works at all — and what LinkedIn says about scraping

The same file opens with a plain warning: "The use of robots or other automated means to access LinkedIn without the express permission of LinkedIn is strictly prohibited." Running a Google search and clicking a result is normal browsing. Pointing a scraper at the results is not, and LinkedIn's User Agreement restricts it too. Search by hand, view profiles normally, and read the terms before you plug in any automation.

12 Ready-to-Run Boolean Search Strings

Every string below uses a senior React developer in Austin as the example. Swap the role, skills and location; keep the structure. Each one has a copy button, a one-line "what this does," and the trap to watch for.

site:linkedin.com/in "react" ("senior software engineer" OR "senior developer" OR "frontend engineer") "austin"

site:linkedin.com/in "EMPLOYER NAME" "react" "austin" -"looking for" -"hiring"

site:linkedin.com/in ("aws certified solutions architect" OR "certified kubernetes administrator") "austin"

site:github.com "austin" ("react" OR "typescript") -site:github.com/orgs

site:github.com "react" "austin" after:2025-06-01

filetype:pdf ("resume" OR "curriculum vitae") "react" "austin" -"job description" -"apply now"

("react developer" OR "frontend engineer") "austin" (site:.dev OR site:.io) -site:linkedin.com -site:indeed.com

site:linkedin.com/in "react" "austin" -recruiter -recruitment -staffing -"talent acquisition" -headhunter

"react developer" "austin" -"apply" -"job description" -"we are hiring" -site:indeed.com -site:glassdoor.com -site:ziprecruiter.com

("speaker" OR "speaking" OR "presented at") ("react conf" OR "reactathon" OR "js conf") "austin"

site:.org ("member directory" OR "find a member") "registered nurse" "texas"

("award" OR "winner" OR "finalist" OR "40 under 40") "engineering" "austin" after:2024-01-01 -site:linkedin.com

Running a lot of these every week? That's usually a sign the sourcing step itself is ready to be systematised — see our guide to recruitment automation for staffing agencies, and for volume roles, the high-volume hiring playbook.

How to Write a Boolean Search String That Actually Works

Copying strings gets you started. Writing your own is what fills the hard job orders. Five habits separate the two.

1. Start from the résumé, not the job description

Clients write job descriptions in their language: "Principal Experience Engineer, Customer Platforms." Candidates write profiles in theirs: "Senior frontend developer, React." Boolean only matches words that are actually on the page, so search the words the people you want would use. Pull three profiles of people you'd happily submit and borrow their vocabulary.

2. Stack synonyms with OR

Since Google dropped ~, nobody is guessing synonyms for you. Build the list yourself: ("frontend engineer" OR "front end developer" OR "UI engineer"). Keep the parentheses — without them, OR can attach to the wrong words and blow the search wide open. And remember the caveat above: on Google, also run the most important variants on their own.

3. Watch the seniority trap

NOT junior feels like a quick way to filter for senior people. It also removes the senior engineer who writes "mentored junior developers" — arguably exactly the person you want. Every exclusion removes a word wherever it appears, not just in the job title. Before you keep an exclusion, run the search with and without it and look at what disappeared.

4. Test, narrow, widen — a four-step loop

1. Run the string and read the whole first page of results.
2. Count how many you'd actually contact. Under a handful? Widen.
3. Change one element — add a synonym, drop an exclusion, loosen the location.
4. Run it again and compare. If you changed three things at once, you'll never know which one helped.

5. Save the string, not just the results

The candidates you find today go stale. The string that found them doesn't. Save working strings against the job order, so the next desk that gets a similar role starts from something proven instead of a blank search box. If you track sourcing performance, it's also how you learn which channels actually produce placements — more on that in recruitment metrics for staffing agencies.

Boolean vs AI Search: Which Do You Actually Need?

This is the question behind the question. 54% of recruitment firms already use automation for candidate search, according to Bullhorn's GRID 2026 report (around 2,300 respondents). So is Boolean a dying skill?

So here's the practical answer: Boolean for sourcing on the open web, semantic search for mining your own database. Most agencies need both.

Think about where the hard part actually sits. On the open web, you need control — the exclusions, the exact credential, the specific competitor. Boolean gives you that. Inside your ATS, the problem is different. You've got years of candidates who were screened for one job order, didn't land it, and never got looked at again — usually because they described themselves in words your string didn't guess. That's a recall problem, and recall is where semantic matching earns its keep.

LinkedIn's own talent resources put the share of workers actively looking at 36% — meaning roughly two-thirds of the workforce isn't actively searching (LinkedIn Talent Solutions). Many of those people are already sitting in an agency database somewhere. A well-built string finds new ones. It rarely finds the ones you already paid to source.

FAQ: Boolean Search for Recruiters