RESPONSIBLE AI

Explainable, defendable scoring you can count on

AI backed by evidence and explained with transparency, because black box LLM scoring isn’t interpretable for high quality hiring.

See how it works
Accurate scoring
Bias audited
Certified

Designed to be explainable, unlike an LLM.

“Strong in patient care” or “weak in critical thinking” is not good enough to explain why a candidate was hired or not.

HiringBranch’s proprietary taxonomy and automatic scoring maps candidate behaviour to specific actions observed during their assessment. This allows the AI to produce job-relevant feedback that can be easily explained.

EXPLAINABILITY SUMMARY

The candidate acknowledged and paraphrased well the customers' issues and offered suitable resolutions. Spoken fluency stayed steady with minor hesitation under pressure. Written follow-up was clear and correctly punctuated.

Observed skills
Paraphrase
Acknowledge
Spoken fluency
Positioning an offer
Writing
Transparent
Know whether a candidate performed successfully and what contributed to that performance.
Structured
Performance is shared in score reports, comparisons, and in plain-language for decision-making and defensibility.
Deterministic
Scoring yields the same result for the same candidate on a test repeat. No silent drift between runs.
EXPLAINABLE SCORING

Score breakdowns

Understand nuance between candidates with controlled, interpretable scoring.

Overall scores & percentiles
Overall scores are comprised of role-required skills and language proficiencies. Know where candidates sit against all candidates globally.
Competency comparisons
Competencies are a weighted collection of role-required skills and the degree to which they’re demonstrated.
Learn more
Language scores
Language scores are broken into speaking, writing, listening and reading demonstrating comprehension and communication fluency.
Observed skills
Observed skills quantifies how often a skill was demonstrated compared to what the role requires.
Learn more

How explainability works

Scoring is controlled and interpretable end-to-end, from the moment a candidate speaks to the moment a recruiter opens the report.

Scored in real time

Scoring happens during the assessment, against the role’s required sub-skills.

Explained instantly

Explainability summaries are generated the moment the assessment ends.

Stored for defensibility

Reports sit inside rich candidate profiles for review, comparison and retention.

From our science team to you

Report: Accurately Measuring Skills Together
Report

Report: Accurately Measuring Skills Together

eBook: CHRO Guide to Skills Verification
Ebook

eBook: CHRO Guide to Skills Verification

Unbiased Hiring Decisions with HiringBranch AI
Report

Unbiased Hiring Decisions with HiringBranch AI

Frequently Asked Questions

What is the difference between inferred skills and verified skills?

An inferred skill is a guess based on proxy signals: a resume keyword, a job title, a confident answer in an unstructured interview, or a personality quiz result. A verified skill is one a candidate has actually demonstrated, in context, against a defined standard, with evidence attached to show how they demonstrated it. HiringBranch only verifies skills, never infers skills.

How does HiringBranch ‘define a global standard’ for scoring with percentiles?

With performance correlations. The HiringBranch Standard is a global benchmark for language, communication, interpersonal and cognitive skills. Every score is validated against real on-the-job performance outcomes. A percentile tells you exactly where a candidate stands in comparison to peers, not just whether they passed.

Is the global percentile for all assessments or individual assessments?

The global percentile is grouped by role and skill competency. You will find percentiles against a candidate’s overall score (compared to every other candidate globally who was assessed for the same role) and competency score (compared to every other candidate in any role who was assessed for that competency e.g. cognitive skills, speaking fluency, etc.).

What does "explainable AI scoring" mean in hiring?

It means every skill score a candidate receives comes with a clear, evidence-based explanation, not just a number. HiringBranch's explainability summary shows the specific parts of a candidate's role play response that drove their score, tied to a defined skill competency, so recruiters always know why a score is what it is.

Do candidates get explanations on their scores?

That is up to the employer. By default, explainability summaries and score breakdowns are built into the candidate profile your hiring team reviews only. Should an employer wish to share results with candidates, that is up to them. 

How is HiringBranch's explainable scoring different from AI tools that use an LLM to judge candidates and then explain the result?

Most "explainable AI" in hiring today is a general-purpose LLM that renders a verdict, then a second AI layer bolted on to write a paragraph justifying it after the fact. HiringBranch's scoring is different: our proprietary and native AI is acoustic- and linguistics-based and purpose-built for skills measurement, so the score was never a black box to begin with. There's no LLM-as-a-judge to explain away, only structured, transparent, deterministic scoring against pre-defined skill competencies.

Does explainable scoring replace a recruiter's judgement?

No, it gives recruiters better information to use their judgement with. Explainability summaries turn a raw score into a plain-language report recruiters can review, compare across candidates, and retain for later, so hiring teams spend their time interviewing proven top candidates instead of guessing at what an AI score actually means. 

Every hiring decision depends on accurate assessments

Get scores you can count on. Book a demo to see explainable scoring in action.