How Verdict reaches a verdict.

Each candidate is read by three analytical lenses, then a separate synthesis weighs them and issues the verdict. The analysis is instructed to disregard the candidate's identity — names, gender, age, and other protected attributes — and, in a head-to-head, the A/B order is randomized, so the score reflects the evidence, not the identity or the order in which candidates were entered.

Every candidate is scored on the same six dimensions — Capability, Track record, Trajectory, Influence, Domain edge, and Risk surface — each from 1 to 5, for a maximum of 30. The rubric is identical for every candidate and every role, which is what makes two evaluations comparable.

Verdict doesn't assert — it quotes. Each judgment is tied to a verbatim snippet from the candidate's CV (or your interview notes), so you can verify the reasoning rather than trust it. No citation, no claim.

The verdict is reached without weighting the candidate's identity, with the entry order randomized. Verdict scores on demonstrated evidence — what someone did and the scope they owned — not on protected attributes. It is a decision-support tool: it surfaces evidence and structures the comparison, but the hiring decision, and the accountability for it, remain yours.

Optionally, you can compile a calibration set from your own past hires and their outcomes (still here, promoted, under-performing, let go). Verdict distills that into an organization-specific profile and applies it as prior context on every future evaluation — so the bar it holds candidates to is yours, and it sharpens the more you use it.