SourcingOS Research

Evidence before recruiting claims.

SourcingOS research is designed around reproducible sourcing questions: what a source actually adds, where a search collapses, when the market is genuinely exhausted, which contact-provider signals are usable, and which AI suggestions recruiters accept or reject.

Published methods and benchmark protocols

Sourcing Lab methods and worked search teardowns

These pages publish the methodology or worked example before any measured product result. A methodology page is not presented as evidence that one provider, source, or search strategy has already won.

Search Teardown · Worked example

Search Teardown: RHEL Administrator Near Annapolis Junction With Secret Clearance

A worked example showing how one natural-language brief becomes explicit requirements, discovery expansions, evidence lanes, verification rules, and calibration questions.

The Sourcing Lab · Methodology

Sourcing Lab Method: How to Compare Contact Data Providers Without Fooling Yourself

A reproducible provider bakeoff for work email, personal email, phone, verification, freshness, overlap, credits, and recruiter-confirmed usability.

The Sourcing Lab · Methodology

Sourcing Lab Method: Measure Which Source Actually Finds People the Others Miss

A controlled source-contribution experiment using common role briefs, independent lanes, identity resolution, evidence-fit retention, duplication, and incremental yield.

Search Teardown · Worked example

Search Teardown: ML Platform Engineer for Production LLM Inference

A worked example for separating model buzzwords from the actual platform, inference, reliability, and production evidence an ML infrastructure role needs.

Search Teardown · Worked example

Search Teardown: Senior Data Engineer for a Streaming Platform

A worked example for sourcing data engineers by system architecture, operational ownership, title adjacency, and open technical evidence instead of a shopping list of tools.

Search Teardown · Worked example

Search Teardown: ISSO / RMF Talent Near Fort Meade With TS/SCI Requirement

A worked cleared-cyber sourcing example that separates RMF evidence, role/title adjacency, geography, federal context, and clearance discovery from authorized verification.

Search Teardown · Worked example

Search Teardown: ICU Registered Nurse With State Licensure Requirement

A worked healthcare sourcing example for specialty evidence, title normalization, location, authoritative licensure verification, and avoiding false equivalence between provider identity and license status.

Worked Talent Map Briefs

These are market-structure hypotheses, not measured candidate-pool counts. They make the title families, donor categories, geography bands, evidence lanes, and scarcity assumptions explicit before SourcingOS reports any live market measurement.

Research program

These are planned research questions, not published benchmark results. SourcingOS will only publish product-derived findings when the sample is large enough, the measurement is meaningful, and the reporting can be aggregated without exposing customer or candidate data.

Research formats

Test the claim.

The Sourcing Lab

Controlled tests of search sources, sourcing techniques, AI workflows, provider/contact quality, and recruiter operating assumptions.

Show the work.

Search Teardown

Role intake → requirements → search plan → queries → sources → false positives → calibration → revised search.

Know where evidence lives.

Source Atlas

Deep field guides to technical, research, professional, federal, licensing, association, and open-web evidence surfaces.

Understand the market before declaring it empty.

Talent Map Briefs

Role, geography, title-family, donor-company, skill, and source-lane intelligence designed to produce an actionable search.

Make good sourcing repeatable.

Recruiter Systems

Operating systems for intake, calibration, Candidate 360, ATS rediscovery, search exhaustion, outreach readiness, and recruiting operations.

Separate capability from marketing.

AI Recruiting Reality Check

Evidence-based analysis of agentic recruiting, AI sourcing claims, data quality, automation boundaries, and recruiter control.

Research rules

Explore SourcingOS Learn