Recruiters · problem and fix
Boolean X-ray search string for recruiters on Mac: how to write and test one
A Boolean X-ray search string for recruiters is one search query that combines AND, OR, NOT, quotes, parentheses and a site: operator to find public profiles through Google. Write it in steps, then test it on result counts before you trust it. This page covers how to write, test and fix one on a Mac, in Chrome and LinkedIn Recruiter.
The short answer
- Use AND, OR, NOT, quotes and parentheses. LinkedIn requires the operators in capital letters.
- For X-ray, start with site:linkedin.com and leave no space after the colon. Then add your job terms.
- Test every string on its result count and on the first page. Narrow it when the list is too big, widen it when too small.
- Move titles, companies and locations into LinkedIn's dedicated filters, where Boolean also works.
- Check your employer's rules before you paste candidate details into any tool, including Lazur.
What boolean and x-ray sourcing strings costs you
Time spent rewriting
Every rewrite costs reading time. You run the string, scan the first page, change a term and run it again. The research did not find a verified hours figure for sourcing. Time one role for a week and you will have your own number.
Candidates you never see
A string that is too narrow hides people who would have answered. You never see them, so the loss does not show up in your count. The 2019 study found that most errors cut recall, which is exactly this kind of loss.
Noise you have to review
A string that is too wide floods the results with the wrong titles and locations. Each false match takes a moment to judge, and those moments add up across a long list. Tight groups and NOT terms cut that pile.
The average cost of a hire
SHRM's 2025 benchmarking puts average cost-per-hire at $5,475 for nonexecutive roles and $35,879 for executive roles. It says executive hires are nearly 7x more expensive. Sourcing is only one part of that cost, so a slow search adds to a bill that is already high.
Trouble with LinkedIn's terms
Copying profiles with a browser plugin can break LinkedIn's User Agreement. That risk can land on you and your employer. Stick to searching and to outreach through channels your company approves.
Why it happens
Strings get typed from memory
Most strings start as a guess. You write what you think a candidate calls their work, then you search. Without a test run, a wrong synonym or a missing word form slips through. You only notice when the list looks thin.
Each tool reads Boolean differently
Google reads a full query. LinkedIn reads Boolean inside filters and ignores stop words in Keywords. Copying one string into the other tool changes what it returns, so the same words give different results.
Titles and skills have many spellings
Candidates write analytics where you wrote analysis, or talent acquisition where you wrote recruiter. Missing synonyms was the most common error type in the 2019 study. The same gap shows up in sourcing.
Long strings break in quiet ways
LinkedIn warns that overly long Boolean queries can cause search result issues. Stop words get dropped, so after sales can match profiles that only say sales. Long strings hide these faults from you.
Terms land in the wrong filter
Job titles typed into Keywords match more loosely than you expect. LinkedIn suggests moving terms into dedicated filter sections. Using the right filter tightens results without extra Boolean.
No record of what worked
Many strings get rebuilt from scratch for each role. Without saved versions and counts, you repeat the same tests. You also lose the good synonyms you found last time.
How to fix it
These work with or without any app.
Use the operators LinkedIn documents
Use AND, OR, NOT, quotes and parentheses in capital letters where LinkedIn expects them. Skip + and minus in LinkedIn Recruiter. The minus sign belongs in Google, where its help page documents it.
Test every string on its result count
Run the string, note the count, then change one group. Write each count next to the string in a sheet. Keep a version only when the count and the first page both look right.
Group synonyms by concept
Put every spelling of one idea in one OR group. For a title, list the formal title, the short form and the common variant. For a skill, list the tool name and the full product name.
Move terms into the right filters
Put job titles in the Job titles filter, company names in Companies, and skills in Keywords. LinkedIn says Boolean works in those filters, so the string gets shorter and easier to read.
Write site: with no space after the colon
Type site:linkedin.com, not site: linkedin.com. Google's help page says to leave no space after the colon. Put the operator in the same search box as your terms, then run it.
Exclude noise on purpose
Use NOT in LinkedIn and a minus sign in Google for words that flood results, such as intern or student. Add one exclusion at a time. Check that it does not remove the people you want.
Split long strings
If a string gets long, move one group into a filter or run two shorter strings. Compare their counts. Short strings are easier to read, edit and explain to a hiring manager.
Read the first page, not only the count
A count can look fine while the top results are wrong. Read at least the first page. Check any candidate details against your company's rules before you share them or paste them anywhere.
Keep a library of strings
Keep a sheet with role, tool, date, string and count. Reuse the title groups for similar roles. Each new role then starts with pieces you have already tested, not a blank box.
Ask Lazur what is on screen
Hold ⌥ ⌘ while the results page is open and ask why a result matches. Lazur answers out loud about what is on screen. Then test the change yourself, because the answer is a guide, not a guarantee.
What a Boolean X-ray string does
Google and LinkedIn read the same words differently. Google Search reads your whole query and returns public web pages. LinkedIn Recruiter reads Boolean inside specific filters, and it does not accept Boolean everywhere. Know which tool you are in before you write a single term.
X-ray is the recruiter name for searching public profiles through a search engine. You type site:linkedin.com followed by your terms into Google. The results show profile titles, which often include a name, a job title and a company. You only see what Google has indexed and what the person has made public.
What the published numbers say
The numbers below come from two sources. One is a 2019 study of search errors in medical reviews. The other is SHRM's 2025 cost-per-hire benchmark. Neither measures recruiter strings directly.
92.7%
Of 137 systematic reviews had at least one search error, in a 2019 study
Salvador-Olivan et al., Errors in search strategies used in systematic reviews78.1%
Of the 137 searches had errors that cut recall, the most common category
Salvador-Olivan et al., Errors in search strategies used in systematic reviewsRead these as context. They show that search errors are common in research and that hires are expensive. They do not show how many candidates your string misses.
Boolean operators you need for sourcing
- AND narrows the results. Every term in the group must appear.
- OR widens the results. Any one term can match, so group synonyms in parentheses.
- NOT removes a term. In Google, a minus sign before a word does the same job.
- Quotes keep a phrase together, such as 'product manager'.
- Parentheses set the order, so (Python OR Go) AND Denver reads the way you mean.
- site: limits results to one domain. Put no space after the colon.
- LinkedIn wants AND, OR and NOT typed in capital letters.
Those seven rules cover most sourcing. You do not need advanced syntax. Google's help page names site:, quotes and a minus sign as the main ways to refine results. Those three do most of the work here.
A worked example you can copy
Say you hire a data analyst in Denver who knows SQL. Build the string in four groups. Each group covers one idea. Keep the groups in parentheses so the order stays clear.
- Title group: ("data analyst" OR "business analyst")
- Skill group: AND (SQL OR "Power BI")
- Place group: AND (Denver OR Colorado)
- Domain: site:linkedin.com at the end, with no space after the colon
Put together, the full string reads: ("data analyst" OR "business analyst") AND (SQL OR "Power BI") AND (Denver OR Colorado) site:linkedin.com. Run it as written first. Then change one group at a time, so you know which change moved the results.
How to test a string before you trust it
A string that looks right can still return the wrong people. Run it, then read the first page of results. Check for the people you want and for the noise you do not want. Record the count after each change, so you can compare versions.
The best published evidence here is indirect. A 2019 study looked at 137 systematic reviews, which are literature reviews in medicine. It found that 92.7% had at least one search error. Of those reviews, 78.1% had errors that cut recall, which means relevant results were missed.
The most common errors were missing synonyms and missing word forms. Those errors match recruiting closely. You search 'analyst' and miss the profile that says 'analytics'. The study covered medical searches, not recruiter strings. No study measures recall for recruiter strings, so treat this as a warning, not proof.
Where Lazur fits in your sourcing routine
Lazur helps most when you are stuck on a term or a filter. It explains what is on screen, points at controls and drafts text. It does not run the search, click buttons or copy profiles for you. You still decide the terms and check the results.
LinkedIn Recruiter limits to plan around
LinkedIn Recruiter supports AND, OR, NOT, parentheses and quoted phrases. It says + and minus are not officially supported, so use AND and NOT instead. Boolean works in the Job titles, Companies and Keywords filters. Project search does not accept Boolean at all.
Two traps catch people. First, stop words in Keywords are ignored. LinkedIn's example is 'after sales', which can return profiles that only contain 'sales'. Second, very long Boolean queries can cause search result problems. Split a long string across the dedicated filters.
Why Google and LinkedIn give different results
Google sees public web pages. LinkedIn Recruiter searches LinkedIn's own member data and filters. A profile can rank high in one and never show in the other. Use X-ray to widen the pool, not to replace your LinkedIn search.
Mistakes that waste your morning
- Typing and, or or not in lowercase in LinkedIn, where capitals are required.
- Using + to force a term in LinkedIn Recruiter.
- Adding a space after site:, which breaks the operator.
- Pasting a whole job description into one string.
- Searching one spelling only, such as recruiter, and missing talent acquisition.
- Putting a job title in Keywords when the Job titles filter fits better.
- Using a browser plugin to copy profiles, which LinkedIn's User Agreement prohibits.
Keep a library of strings you have tested
Save each string with its role, the tool, the date and the result count. Next time you open a similar role, start from a version that already worked. A Google Sheet is enough. Select the header row and use Data > Create a filter to sort by role or by count.
Fixing a string that returns too many or too few
| Symptom | Likely cause | Fix |
|---|---|---|
| Too many profiles from other fields | Synonyms too broad, such as 'analyst' alone | Add a must-have skill with AND, and exclude the unwanted field with NOT |
| Too few profiles | Too many AND groups | Drop the least important group, or move that term to an OR group |
| Good profiles missing | Missing synonyms or word forms | Add the variant in quotes, such as "analytics" |
| Results ignore a word | Stop word in Keywords | Move the term into the Job titles filter |
Change one thing at a time. If you change three groups and the count moves, you cannot tell which change did it. Test, record, then adjust again.
Candidate data and your employer's rules
A search string holds no candidate data. The profiles you read do. Lazur listens and looks at the screen only while you hold the shortcut. When you ask, audio and a screenshot go to AI providers to answer, and they are not stored. Content is never used to train AI models.
Your employer sets the rules on candidate data, not this page. Check them before you paste profile details into any tool, including an AI assistant or a shared sheet. Regulated hiring, such as roles that need licensed nurses or lawyers, often has stricter rules still.
A short routine for your next role
Start with the title and skill groups. Those two groups do the most work. Add location only after the first run looks sensible. Then move terms into LinkedIn's filters and save the version that worked.
Finish by writing one line on what you would change next time. The next role starts from that line, not from a blank box. Over a few roles, your library becomes the fastest part of sourcing.
Faster with Lazur
Where Lazur helps with boolean and x-ray sourcing strings
Hold ⌥ ⌘ and say what you want. Lazur sees your Chrome window. Lazur draws a circle and a label around the button or menu you need.
- 1
Hold ⌥⌘ and talk
- 2
Lazur draws a circle and a label around the button or menu you need.
- 3
You stay in Chrome. Nothing to copy into a chat window.
More things to say in Chrome
- “What does this OR group match on this page?”
- “Where is the Job titles filter?”
- “Walk me through adding a Keywords filter.”
- “Draft a title group for a data analyst role.”
- “Hold fn and dictate my notes into the sheet.”
- “Research how LinkedIn treats stop words in Keywords.”
Free for 3 days, then $18/month. Download for Mac
You say“Lazur, where is the Job titles filter?”
Lazur · points at itLazur circles the Job titles filter and labels it. You type the title group yourself and run the search.
What recruiters say to Lazur for this
Each one is a single hold of ⌥ ⌘.
“What does this OR group match on this page?”
Hold ⌥ ⌘ while the results page is open, and Lazur answers out loud about what it sees.
Explains what's on screen
“Where is the Job titles filter?”
Lazur draws a circle and a label around the filter, so you find it without hunting.
Points at it
“Walk me through adding a Keywords filter.”
Lazur guides you one click at a time and waits for each click. You do the clicking.
Walks you through it
“Draft a title group for a data analyst role.”
Lazur drafts the text. You test the string on result counts before you use it.
Drafts it
“Hold fn and dictate my notes into the sheet.”
Dictation writes clean text at the cursor in any app. Hold fn to start and let go to finish.
Types what you say
“Research how LinkedIn treats stop words in Keywords.”
Research runs in Lazur's own Chrome tabs and returns a summary and a report. Check the sources it cites before you rely on them.
Researches in the background
A plan you can start this week
- Day 1
Write the core groups
Pick one open role. Write the title group and the must-have skill group with OR and parentheses. Leave location for later. Run the string once and record the count and the first page's titles.
- Day 2
Add location and exclusions
Add the place group, then one NOT term for the noise you saw on Day 1. Run the string after each change. Write the count beside each version, so you can see which change moved the results.
- Day 3
Test on the first page
Read the first page for the wrong titles and the wrong places. Adjust one group, then rerun. Stop when the first page looks right and the count is workable for the role.
- Day 4
Move terms into LinkedIn filters
Open LinkedIn Recruiter. Move titles into Job titles, companies into Companies and skills into Keywords. Check Keywords for stop words such as after, which LinkedIn ignores.
- Day 5
Save and note the lesson
Save the final string, the count and the date in a sheet. Write one line on what you would change next time. The next role starts from that line.
Questions people ask
Is X-ray searching the same as scraping LinkedIn?
No, but the line can blur. Searching Google with site:linkedin.com is a search. LinkedIn's User Agreement bars software, including browser plugins, that scrapes or copies profiles. Check your company's policy and LinkedIn's terms. This is not legal advice.
Can I use + in LinkedIn Recruiter?
LinkedIn says + and minus are not officially supported. Use AND to require a term and NOT to exclude one. Type both in capital letters.
Does Boolean work in LinkedIn Project search?
No. LinkedIn says Project does not support Boolean search. Use the Job titles, Companies and Keywords filters, where Boolean does work.
Why does my string return profiles that only match part of it?
Stop words in the Keywords filter are ignored. LinkedIn's example is 'after sales', which can return profiles that only contain 'sales'. Move the phrase into a filter or quote it.
How long can a Boolean string be?
LinkedIn warns that overly long Boolean queries can cause search result issues. Google's help page does not state a length limit. Split long strings into shorter ones and compare their counts.
Can Lazur write my Boolean string?
Lazur drafts text when you ask. It explains operators and points at filters on screen. It does not run the search, click buttons or copy profiles for you. Test every draft on the result count before you use it.
Is there a Mac app that builds X-ray strings?
The research did not find a Mac-native Boolean app. Most tools are web tools or Chrome extensions. Check any extension's terms and its macOS support before you install it.
Sources
- Use Boolean to filter search results in Recruiter and Recruiter Lite · LinkedIn Help · checked 2026-11-09
- Refine Google searches · Google Search Help · checked 2026-11-09
- Errors in search strategies used in systematic reviews and their effects on information retrieval · Journal of the Medical Library Association (via PubMed Central) · checked 2026-11-09
- SHRM Releases 2025 Benchmarking Reports: How Does Your Organization Compare? · SHRM · checked 2026-11-09
- LinkedIn User Agreement · LinkedIn · checked 2026-11-09
Fix boolean and x-ray sourcing strings with Lazur, free for 3 days
Hold a key and ask. Lazur sees your screen, points at the button and talks you through it, in any Mac app. No credit card.