The Search Brief Standard: When a Role Is Sourceable

This standard is for in-house recruiters, sourcers, talent leaders, and founders who write their own briefs and want to stop guessing whether a role is ready to work. It gives you fixed pass/fail criteria for grading a search brief, so two people reviewing the same brief reach the same verdict, and two sourcers working from it build substantially the same shortlist. Adopt it as team policy and you replace "the brief feels vague" with a line anyone can apply.

Existing intake checklists tell you what topics to cover. None of them tell you when a brief is done. That gap is expensive: a brief that reads fine but leaves interpretation open produces three sourcers chasing three different candidates, and the drift only surfaces weeks later at shortlist review. This document defines done.

What makes a brief sourceable, and how it differs from a job description

A brief is sourceable when every must-have is tied to observable profile evidence, the criteria are scored on a shared scale, and the whole thing predicts convergence: two independent sourcers would find similar people. That last property is the standard. Everything else is a means to it.

The line against a job description is directional. A job description is written to attract applicants; a brief is written to direct a search. A JD says "we're looking for a strategic leader with 10 or more years of experience." A brief says something a sourcer can act on: "scaled a SaaS sales team from 8 to 30 reps between $8M and $35M ARR, in a product-led motion where outbound is 40% of pipeline, in the last four years." The first reads well to a candidate. The second tells a sourcer exactly who to find.

Named practitioner frameworks converge on six or seven fields. Metaview's hiring brief lists six elements: the outcome, must-haves and their evidence, trade-offs, constraints, calibration profiles, and an honest pitch. Majhi Group's search brief runs seven sections: company context, why this role and why now, role scope and authority, a success definition at 30, 90, and 365 days, the ideal candidate profile, compensation structure, and interview process. The overlap is the signal. If your brief is missing outcomes, the must-have split, evidence, or calibration, it is not sourceable yet.

The required fields: what a passing brief must contain

A passing brief contains six load-bearing fields, each with its own acceptance bar. Miss one and the brief fails, no matter how polished the rest reads.

• Outcomes. Three to eight ranked, measurable results the hire must deliver. The Who method (Geoff Smart and Randy Street) frames the scorecard as mission, outcomes, and competencies, with most roles carrying three to eight outcomes ranked by importance.
• Must-haves with evidence. Each must-have names the observable profile signal that proves it. No adjectives standing alone.
• Nice-to-haves. Everything that failed the veto test, kept visible so a sourcer knows what strengthens a candidate without gating them.
• Constraints and trade-offs. Comp range, location, timeline, and the honest hard parts of the role. Contradictions here must surface at intake.
• Calibration profiles. Real example profiles both parties agree represent a strong candidate.
• The scoring rubric. Five to seven criteria on a 1-to-5 scale with worked examples of weak, acceptable, and strong.

The most common failure is treating the must-have section as a wish list. The fix is a single test, covered next.

The veto test: separating a must-have from a preference

A must-have is any requirement where you would immediately disqualify a candidate who lacks it. Everything else is a preference. That is the entire distinction, and it is testable in one question.

Workable frames it directly: ask the hiring manager "should we immediately disqualify candidates who don't have X in their resumes?" Bonsai adds probing follow-ups: "Is X certification truly required?" and "Could we consider candidates from Y industry?" If the honest answer softens, the criterion moves to nice-to-have. The point is to force preferences out of the veto tier, because every false must-have narrows the funnel for no defensible reason.

One class of veto is not just weak but risky. Rejecting a candidate for being "overqualified" carries EEOC exposure, because excess qualifications often correlate with age, and refusing to hire on that basis has been treated by US courts as possible age-discrimination code. A veto must be job-related and evidence-based, or it does not belong in the brief.

Converting a must-have into a searchable, evidence-backed criterion

A criterion is searchable only when it names the observable evidence a profile would show, not a label a human would infer. Separate sourcing evidence from interview evidence: a profile may show relevant experience, but it cannot prove judgment, communication, or how someone handled a difficult situation.

The Recruiter's Handbook demands observable evidence over labels. Instead of rating someone on "leadership," define what leadership evidence looks like: led a team of 10 or more through rapid growth, improved retention, built a hiring plan, and delivered a measurable operating result. Each of those is visible on a profile or in a portfolio. "Leadership" alone is not.

This is where two sourcers quietly diverge. Give them "strong communicator" and one filters on conference talks, another on published writing, a third ignores it because nothing on a profile proves it. Give them "led a team of 10+ through headcount growth with a documented hiring plan" and they filter on the same thing.

Watch the skill-string trap. In Refolk's index of professional profiles, only 20 US recruiters and sourcers list "Boolean Search" as a skill, against 986 who list "LinkedIn Recruiter" - a 49x gap. Sourcers treat Boolean as an assumed baseline, not a badge, so a brief that filters on the literal presence of a skill string misses almost everyone who has it. Filter on titles and tool skills, not self-reported skill labels.

That table is a warning about your own filters. The same logic applies to any must-have expressed as a skill tag: the label is present on far fewer profiles than the capability, so gate on demonstrable work instead.

Writing evidence-backed criteria by hand is slow, and matching them against real profiles is slower. This is the friction Refolk removes: you describe the must-have in plain English, name the evidence, and get back the people whose public work actually shows it, so you can test whether your criterion returns a sensible pool before you commit it to the brief.

The rubric: how many criteria, what scale, and where the pass line sits

A passing rubric has five to seven criteria, a 1-to-5 scale, and worked examples of what weak, acceptable, and strong look like for each. That is enough structure for a stranger to score a profile and reach the same number a colleague would.

The named sources agree closely on sizing. Pin uses must-have, nice-to-have, and bonus tiers, each competency on a 1-to-5 scale with concrete examples of what a 3 looks like versus a 5. The Recruiter's Handbook says five to seven criteria are usually enough, with agreed definitions of strong, acceptable, and weak evidence. Bonsai runs six to eight screening items on a 1-to-5 scale.

The step most briefs skip is the pass line. A rubric without a threshold still leaves "close enough" to the reader. The UN system fixes this bluntly: every applicant has to pass an 80% threshold of satisfactory answers for a job opening, or the application is not eligible for further review. You do not need that exact number, but you need a number - for example, all must-haves plus at least two of four nice-to-haves. A fixed line converts subjective judgment into a rule two reviewers apply identically.

The procedure: from intake to a signed, sourceable brief

Run these eight steps in order. Each has a "done" condition, so you can stop and grade progress at any point rather than discovering at the end that a field is missing.

The calibration step in detail

Calibration is the step that actually proves convergence, so do not skip it. Scede describes sourcing example profiles and presenting them, alongside qualifying questions, back to the hiring manager to sharpen shared understanding. Pin describes a dry run: everyone scores the same one or two anonymized past candidates against the rubric, reveals the scores, and discusses the outliers. The point is not to reach the same score on the dry-run candidate. The point is to surface where one reviewer's "3" means what another's "5" means. RecruitBPM adds reviewing sample resumes to align on what strong, average, and weak look like against the requirements.

If scores diverge on the dry run, the rubric language is undefined and the brief is not done, regardless of how complete the fields look. One Scede case had a scaleup role sit open six months across three recruiters before calibration finally aligned the search.

How this goes wrong: failure modes and false positives

Most bad briefs pass a casual read and fail in the field. These are the specific ways it happens and the check that catches each one. This section is the working core of the standard: grade against it, not against your gut.

• Adjective masquerading as a must-have. "Strong leadership" looks fine but no profile signal proves it. Check: every must-have names observable evidence, as in the Handbook's "led a team of 10+" example. False positive: a criterion two sourcers each interpret differently.
• Interview-only trait listed as a sourcing must-have. "Great communicator" cannot be sourced. Check: if a profile cannot show it, it belongs in the scorecard, not the search filter.
• Preference smuggled in as a veto. A manager later rejects a strong candidate over a nice-to-have. Check: apply the disqualify question to every must-have; if the answer is no, it is a preference.
• Illegitimate veto. "Overqualified" and age-proxy criteria create EEOC exposure and are not defensible. Check: confirm each veto is job-related and evidence-based.
• Hidden criteria added mid-search. A new requirement appears at shortlist review and invalidates prior sourcing. Check: a versioned brief with a change log; if a criterion is not in the current version, it cannot reject a candidate.
• Uncalibrated rubric. Two reviewers score the same profile 3 versus 5. Check: run the anonymized dry run before real candidates; divergence means the language is undefined.
• Unrealistic brief. A specialist role in a tight market with below-market comp, no relocation, and a two-week timeline. Check: the trade-off contradiction must surface at intake, not weeks in.
• No sign-off. A silent no-vote stakeholder reopens settled criteria late. Check: all no-vote decision-makers sign before outreach.

There is one genuine disagreement in the sources worth naming. Some frameworks say lock and sign the brief before outreach; others say let it evolve with market feedback. The resolution is order: lock the must-haves and get sign-off first, then treat the brief as a versioned source of truth that you update deliberately with a change log. Locking prevents silent reopening; versioning prevents the brief from going stale. They are not in conflict once you sequence them.

Watch the market before you reuse a brief across geographies

Do not reuse a brief across markets without re-checking the title and the available pool, because supply and title meaning shift geographically. A criterion that surfaces the right people in one market can surface the wrong profession in another.

The clearest example is sourcer supply itself. In Refolk's index, US sourcer-type profiles outnumber the UK by roughly 11 to 1.

The gap is not only volume. The UK sample skews toward "Sourcing Specialist" in procurement and supply chain rather than talent, so a brief written for a US talent sourcer, reused in the UK on the same title filter, surfaces the wrong profession. Re-scope the title, then re-check the pool size, before you assume a brief travels.

number: 4,923
label: US sourcer-type profiles in Refolk's index
note: Against 437 in the UK, an 11.3x gap that changes how aggressively you can filter in each market.