Free tool
In buildGet a boolean string you can actually edit
Enter a role, the skills that matter and a location, and get search strings shaped for each major platform rather than one string that half works everywhere.
The Surhires boolean string generator will take a role title, a skills list and a location, and produce search strings adapted to the syntax each major platform accepts. It expands job title synonyms, groups skill alternatives with OR, and excludes obvious noise. A generated string is a starting point you edit, not a finished search.
By Surhires Editorial · Published · Reviewed
In the product: in build for Wave 1 as a free public tool
The operators, stated properly, because most guides get them wrong
Boolean search has five things in it and they combine predictably. AND narrows: both terms must appear, and on most platforms a space between two words already means AND, which is why adding it explicitly changes nothing. OR widens: either term will do, and it is the operator that does the real work in recruitment because job titles are not standardised. NOT excludes, written as NOT or a minus sign depending on the platform.
Quotation marks make a phrase, so "project manager" matches the two words adjacent and in that order rather than a document containing both words anywhere. Parentheses group, and they are the difference between a search that works and one that quietly returns nonsense: nurse AND (paediatric OR pediatric OR children) is a different query from nurse AND paediatric OR pediatric OR children, and the second one returns every document mentioning children.
Wildcards truncate a stem. Nurs* catches nurse, nurses and nursing on platforms that support it. Support is inconsistent, and a wildcard that is not supported is usually treated as a literal asterisk rather than as an error, which means it silently returns nothing.
- AND narrows the result set; on most platforms a space already implies it
- OR widens, and carries most of the value because titles are not standardised
- NOT or a leading minus excludes; check which form the platform accepts
- Quotation marks bind a phrase into a single adjacent term
- Parentheses control the order of evaluation and are not optional in a mixed query
- Wildcards truncate a word stem, but support varies and failure is usually silent
Every platform reads a string slightly differently
A string that runs on one search surface will not necessarily run on another, and the failures are quiet rather than loud. Some platforms cap the number of operators or the total character length and truncate the rest. Some ignore parentheses beyond a nesting depth. Some treat a hyphenated term as an exclusion because the hyphen reads as a minus. Some strip wildcards. Some have their own field syntax, a title or company qualifier, that does the job better than a general keyword search.
The generator produces a variant per platform rather than one universal string, and tells you what it changed and why. That is more useful than a single string plus a disclaimer, because the differences are the whole problem. It also flags where a platform's own structured filters, location radius, seniority, current employer, will do a cleaner job than cramming the same constraint into the keyword box.
What you enter and how the string gets built
You give it a role title, the skills that genuinely matter, a location, and optionally a seniority and things to exclude. It expands the title into a synonym group, because the person you want may be a software engineer, a developer, a programmer or a member of technical staff depending on who wrote their job title. It groups the skills you marked essential with AND and the interchangeable ones with OR.
Then it adds the exclusions that save you time on most searches: recruitment agency terms if you are searching a public web index, obvious student and internship signals if you asked for senior, and the noise terms that come with the specific title you entered. Every one of those additions is shown separately so you can strip the ones you disagree with.
A generated string is a first draft, and treating it as an answer is the mistake
This is the honest part. No generator knows your market. It does not know that in your city the role is called something else, that a particular former employer is a reliable source of the skill you need, that the certification you listed was renamed in 2023 and both names are still in use, or that the term you excluded is also how half your best candidates describe themselves.
So the workflow that works is to run the generated string, look at the first twenty results, and then edit. If the results are dominated by one wrong role family, add an exclusion. If the count is in the tens of thousands, add an AND term. If it is in single figures, take the most restrictive term out and see what surfaces. Two or three edits usually gets you somewhere a first-pass string never would.
The value of the generator is that it removes the syntax errors and the blank-page problem. It does not remove the sourcing judgement, and any tool that claims to has not been used on a hard requisition.
What happens to what you type into this tool
The generator will process the role, skills and location you enter to build the strings and will not retain them. No account is needed and none is created by using it. The inputs here are role and skill terms rather than personal data, which makes this the lowest-risk tool on the page.
That changes if you paste in something else. Recruiters sometimes paste a candidate profile or a resume into a string builder to extract search terms from it. If you do that, you have put another person's personal data into a free web tool, and that is a processing decision you are making as controller under your own privacy notice. It is not one we can make for you. Extracting the terms by hand takes a minute and avoids the question entirely.
What the generator will not do
It will not run the search. It produces text you paste into a platform you have your own access to, under that platform's terms. It will not scrape results, bypass a login or automate anything against a site that does not permit it, and it will not return candidate profiles.
It will not know how many results a string returns, because it has not run it. It will not guarantee a string is accepted, because platforms change their syntax without notice and a string that worked last quarter can silently degrade. It will not replace a platform's own structured filters, and where those are better it says so rather than pretending the keyword box is the only route.
Sourcing your own database before you search anywhere else
The cheapest candidate you will ever place is one already in your database, and boolean strings for external platforms are the second step rather than the first. Inside Surhires the Candidate Sourcing Agent searches your own records against a live requisition before you go outside, surfacing people who already match and who you have already spoken to.
That search runs against structured fields, normalised skills, seniority, location, notice period, alongside the parsed resume text, so it can express a brief that a keyword string cannot. Saved searches turn a brief into a live list that repopulates as new candidates arrive. The boolean generator remains useful for external platforms; it just should not be the first place you look.
What you get
Per-platform variants
A separate string shaped for each major platform rather than one string that half works.
Syntax differences named
Says what it changed for each platform and why the change was needed.
Title synonym expansion
Builds an OR group across the ways the same role gets titled in the market.
Essential versus interchangeable
Skills you mark essential join with AND; alternatives group with OR.
Noise exclusions
Adds the usual agency, student and wrong-family exclusions, each shown separately.
Editable output
Every added term is listed so you can strip the ones you disagree with.
Operator reference
AND, OR, NOT, quotes, parentheses and wildcards explained beside the output.
Length warnings
Flags where a string exceeds the operator or character limit a platform accepts.
Wildcard caution
Marks platforms where an unsupported wildcard fails silently rather than erroring.
Filter suggestions
Points at a platform's structured filters where they beat a keyword clause.
No retention
Role, skills and location are processed to build the strings and are not kept.
No scraping
Produces text you paste yourself. It does not run searches or collect results.
Questions recruiters ask
Why do I need parentheses?
Because without them the platform decides the order of evaluation and usually decides wrongly. Nurse AND (paediatric OR pediatric OR children) asks for a nurse with one of three qualifiers. The same terms without brackets can be read as nurse and paediatric, or pediatric alone, or children alone, which returns every document mentioning children. Group anything with an OR in it.
Will the generated string work on every platform?
No, which is why the tool produces a variant per platform instead of one universal string. Platforms differ on operator limits, character caps, nesting depth, hyphen handling and wildcard support, and most failures are silent rather than an error message. The generator tells you what it changed for each target so you can see the difference.
Should I use the string exactly as generated?
No. Run it, look at the first twenty results, and edit. If one wrong role family dominates, add an exclusion. If the count is enormous, add an AND term. If it is tiny, remove your most restrictive clause. A generator removes syntax errors and the blank page; it does not know your market, and pretending otherwise wastes a sourcing session.
Do wildcards always work?
No, and that is the trap. Support varies by platform, and an unsupported wildcard is generally treated as a literal character rather than rejected, so the search returns nothing and looks like a genuine empty result. The generator marks the platforms where this happens and offers an explicit OR group as the safer alternative.
Do you store what I type in?
No. Role, skills and location are processed to build the strings and are not retained, and no account is needed or created. If you paste a candidate profile or a resume in to pull search terms from it, you have added somebody else's personal data to a free web tool, and that decision is yours as controller rather than ours.
Does the tool run the search or return candidates?
Neither. It produces text you paste into a platform you already have your own access to, under that platform's own terms. It does not scrape, does not bypass a login, does not automate against any site and returns no profiles. Anything that offered to do those things would be creating a problem for you rather than solving one.
Is it live yet?
No. It is in build for the Wave 1 release as a free public tool, and there is no published launch date. This page sets out what it will do when it arrives. In the meantime the operator reference above is the useful part, and it works whether or not our generator exists.
Keep reading
- Rediscover the candidates you already own before you buy more
- Source remote candidates who can work your hours
- Capture from LinkedIn today, connect Recruiter when access is granted
- The lists you rebuild every month, built once
- One candidate record that actually stays current
- One-click capture of the profile in front of you
See it against your own reqs
Bring one live role and three resumes. In twenty minutes you will see the match scores, the shortlist and the placement invoice that comes out the other end.