Boolean Sourcing Search String
When to use it
Turn a role's requirements into search strings for LinkedIn Recruiter or a Google x-ray, with the reasoning shown.
The prompt
I'm sourcing candidates for a {{Job Title}} role. Build me search strings, and explain them. MUST-HAVE SKILLS: {{list}} NICE-TO-HAVE SKILLS: {{list}} JOB TITLES TO INCLUDE AS SYNONYMS: {{list, or suggest some}} LOCATION: {{location, or "remote in {{country}}"}} SENIORITY: {{level}} WHO I DON'T WANT: {{e.g. agency recruiters, salespeople whose profiles match on the tool name, students}} Give me: 1. A LinkedIn Recruiter Boolean string. Respect its actual limits — no wildcards mid-word, mind the character cap, and remember it searches the whole profile rather than just the current title. 2. A Google x-ray string for site:linkedin.com/in. 3. A deliberately narrow and a deliberately broad version of each, so I can start tight and loosen it as results dry up. 4. Synonyms and alternate spellings for every skill — "Postgres" OR "PostgreSQL", "K8s" OR "Kubernetes", British and American spellings. 5. Exclusion terms that cut the predictable false positives, and the risk of each: tell me where an exclusion also removes people I want. 6. Your reasoning for anything left out as too broad or too narrow. 7. What this string will systematically miss — the qualified person whose profile simply doesn't use these words. 8. An adjacent-industry version that targets people from a different sector or a nearby job family doing substantially the same work, with one line per swap on what that group would have to learn on the job. Put each string on a single line I can copy without reformatting.
Tip
The 'what will this miss' answer is the valuable one. A string that returns 40 clean results has usually excluded a few hundred people who describe the same work in different words.
How to use it
Paste the prompt into ChatGPT, Claude or whichever assistant you use, then replace every {{bracketed}} part with your own detail. The more specific and messier your input, the better the output — a model given raw notes has more to work with than one given a tidy summary you wrote first.
More Sourcing prompts
Open the full tool for all 87 prompts, the glossary, and the JD decoder.