01
The workflow
Employers design job-specific assessments. Candidates demonstrate traits through evidence and judgment. Honrly scores what was shown-and whether the response appears authentic. The sequence is deliberate: competency design first, then controlled question generation, then evidence scoring, then integrity verification. AI accelerates structured design; it does not invent undefendable personality labels.
- Select competencies and assign weights for a role
- Generate controlled behavioral and situational questions from validated templates
- Freeze the assessment version so comparable candidates get an equivalent experience
- Collect text, audio, or video responses
- Score against predefined rubrics-not a generic LLM opinion
- Verify integrity and produce an explainable report for human review
02
What Honrly actually scores
For each answer, Honrly evaluates observable evidence-not whether the response “sounds adaptable.” Scoring is tied to behaviorally anchored rubrics so reviewers can see why a competency score moved up or down. The goal is job-related evidence you can explain to a hiring manager, a candidate, or an auditor-not a mysterious fit score.
- Evidence - Did the candidate provide a real, relevant example?
- Ownership - Did they take personal responsibility or speak vaguely about “the team”?
- Judgment - Were decisions appropriate to the situation?
- Specificity - Actions, constraints, and outcomes explained?
- Reflection - Learning or what they would change?
- Competency evidence - Observable behaviors supporting the selected competency
- Consistency - Do examples across questions support or contradict one another?
- Response integrity - Suspicious assistance, copying, or impersonation signals?
03
Not another Thomas-style self-report
Traditional psychometric tools often ask candidates to describe themselves through standardized items. That can be useful for broad preference mapping, but self-report is easy to game and hard to defend as demonstrated job evidence. Honrly asks candidates to demonstrate a trait through evidence, judgment, and lived examples-then verifies those responses are theirs. If you are comparing categories, think verified behavioral assessments plus integrity-not another personality battery with an AI wrapper.
- Open-ended evidence interviews instead of forced-choice self-description as the primary signal
- Rubrics tied to observable workplace behaviors
- Authenticity checks so AI-written fluency is not mistaken for competence
04
Explainable candidate reports
Hiring teams receive role alignment, competency-by-competency scores, evidence supporting each score, missing or contradictory evidence, confidence, integrity observations, and suggested follow-up questions. Honrly does not automatically make the final hiring decision. Reports are designed for human review: what was demonstrated, what was missing, and whether authenticity concerns should change how the evidence is weighted.
- Competency scores with supporting evidence excerpts
- Gaps and contradictions surfaced for interviewer follow-up
- Integrity observations shown alongside content quality-not collapsed into one opaque number
- Human override and notes as part of the decision record