AI Candidate Scorecards: What They Are and Why They Matter
A candidate scorecard is a structured, explainable summary of how a candidate performed against the criteria for a specific role. Instead of a single opaque rating, a good scorecard breaks the evaluation into named dimensions (skills, experience, role fit) with a score and a stated rationale for each, plus supporting evidence from the application and screening. Scorecards make shortlist decisions faster (hiring managers see signal, not raw applications), more consistent (every candidate is judged on the same dimensions), and more defensible (each decision is documented and auditable).
What a strong scorecard contains
- Named evaluation dimensions tied to the role, not a single overall number
- A score and a written rationale for each dimension
- Supporting evidence pulled from the application and the screening
- A consistent structure across every candidate for the same role
Why explainability is the point
A score no one can interrogate does not get trusted, and does not survive scrutiny. An explainable scorecard lets a hiring manager see why a candidate ranked where they did, and lets a compliance team show that decisions were made consistently and for stated reasons.
This is why explainability is moving from a nice-to-have to a requirement: bias-audit and transparency rules for automated hiring expect a documented, defensible basis for decisions.
How Eximius generates scorecards
Eximius scores every candidate against the role’s criteria and produces a scorecard with a rationale per dimension, drawing on both the parsed application and the structured screening (chat, voice, or video). The result is a shortlist hiring managers can act on and defend.
Frequently asked questions
What is the difference between a scorecard and a rating?
A rating is a single number. A scorecard breaks the evaluation into named dimensions with a rationale for each, plus supporting evidence, which is what makes it explainable and defensible.
Are AI scorecards biased?
Any evaluation can carry bias if built carelessly. The safeguards are consistent, role-specific criteria applied to every candidate, explainable rationales, human oversight, and bias auditing: the same practices required by regulations like Local Law 144.