Standards for evidence, fairness, and recruiter review.
Every assessment and report should help recruitment teams make clearer decisions while avoiding exaggerated claims about candidates, roles, or outcomes.
The standards we apply.
These rules keep the platform focused on responsible decision support for agencies.
Evidence before labels
Reports should explain what evidence supports a statement instead of relying on vague candidate labels.
Limitations in plain language
Recruiters should see what the assessment can and cannot support.
Interview verification
Risk areas should become questions to verify, not silent reasons to reject a candidate.
Publication criteria.
Before a workflow or report language is used, it should pass practical checks.
Required
- Clear construct definition.
- Candidate-friendly wording.
- Report language that avoids automatic decisions.
- Role-relevant interpretation.
- Human-review disclaimer.
Not allowed
- Clinical diagnosis language.
- Automatic reject/advance logic presented as final.
- Unsupported certainty about future performance.
- Hidden assumptions that recruiters cannot review.
- Client-facing claims that remove human responsibility.
Fairness and consistency.
Recruitment agencies need a repeatable structure so candidates are compared more consistently.
| Standard | Agency benefit |
|---|---|
| Consistent prompts | Candidates are assessed through the same workflow for the same role. |
| Transparent bands | Recruiters can see what a signal means instead of guessing from a raw number. |
| Risk verification | Concerns become interview topics rather than hidden decisions. |
| Human review | The agency remains responsible for final interpretation and client advice. |
Common questions.
Direct answers to how these standards get applied in practice.
Do these standards apply to every language version?
Yes — the same evidence, fairness, and human-review standards apply across all 9 language versions, not just the English original.
What happens if a report is found to violate a standard?
Report language and workflows are reviewed against these publication criteria before use; anything using disallowed language like automatic reject/advance logic is corrected rather than published as-is.
Do these standards ever restrict what an assessment can measure?
No — they govern how results are worded and used, not what traits or competencies an assessment is built to measure.
Who is responsible for the final hiring decision under these standards?
The agency and its recruiters — the standards specifically require that human review, not the software, carries that responsibility.
Related pages.
For more detail, review how results are calculated and how validation is handled.