Scoring Consistency

Every interviewer scores the same way

Scoring consistency means two interviewers who hear the same answer give it the same rating. Ratio puts the same four RATE prompts under every question, one four-point scale set to the level of the role, and one fixed rule for combining ratings. People rate; AI never scores a candidate.

Book a DemoSee a sample interview kit

What every interviewer listens for

RATE gives every question the same four lenses. Interviewers use them to judge the answer, then give it one rating.

  • Relevance

    Specific, real detail, not a general answer.

  • Approach

    How they did the work, in their own words.

  • Trade-offs

    What made it hard, and what they gave up.

  • Evidence

    What happened, and how they know it worked.

How RATE works

1

One simple rating per answer

Every answer gets one rating on the same four-point scale: unable to answer, below, meets, exceeds. There is no middle score to hide in, so the interviewer has to decide whether the answer met the bar.

A hiring-manager question for Month-End Close, and the interviewer's rating of the answer: three of four stars, 'Meets', with an evidence note quoting what the candidate saidHiring-manager interview · Julian ZimmermanQuestion 1 of 8End interviewMust-haveDomain KnowledgeAdvancedSituationalDay three close is in two hours, but an unmapped variance is off by fifty thousanddollars, and the plant manager won't adjust their ledger. Do you post the packagewith the discrepancy or delay reporting?Scoring rubric · Month-End Close · AdvancedUnable to answerThe candidate could not give an answer to assessBelowFalls short of what the role needs at this levelMeetsMeets what the role needs at this levelExceedsGoes beyond what the role needs at this levelYour ratingClear1 Unable to answer · 2 Below · 3 Meets · 4 Exceeds3 of 4 · MeetsMeets what the role needs at this level.EVIDENCE NOTE“Named the entry she would trace first andposted with a disclosed variance. Cost ofthe delay only came out on the follow-up.”Up next2Internal Controls3Financial ReportingNext question
  • Four RATE prompts under every question say what the interviewer is listening for
  • Four prepared follow-ups, one per lens, for when an answer misses one
  • A one-line evidence note travels with every rating
2

Scored at the role's level

A senior accountant and a junior one can both answer a close question correctly. Ratio sets each skill's proficiency from the job description, so the question gets harder with the level while the four lenses stay the same. "Meets" means meets for this role, and every interviewer holds the same bar.

Month-End Close, a must-have, in both rounds: the screening question with Elena Marsh and the harder hiring-manager question with Devon Okafor, each with the same four RATE anchorsDefine the jobReview skillsInterview stages4QuestionsEMMonth-End CloseDomain KnowledgeAdvancedScreening round30 min · Elena MarshSituationalIf a subledger doesn't match the general ledger hours beforemonth-end close, how would you handle it?Did the candidate:RRELEVANCEGive specific detailAAPPROACHExplain their approachTTRADE-OFFSDiscuss the tradeoffEEVIDENCEDescribe what good looks likeHiring-manager round45 min · Devon OkaforSituationalDay three close is in two hours, but an unmapped variance is offby fifty thousand dollars, and the plant manager won't adjusttheir ledger. Do you post the package with the discrepancy ordelay reporting?Did the candidate:RRELEVANCEGive specific detail, not just claiming they managed theclose while only compiling provided spreadsheetsAAPPROACHExplain their approachTTRADE-OFFSDiscuss the tradeoffEEVIDENCEDescribe what good looks like
  • A proficiency level per skill: beginner, intermediate or advanced
  • Screening questions ask the plain version; hiring-manager questions add a real constraint
  • One rating per answer, relative to the level, not to the other candidates
3

Must-haves count more

Must-have skills count more than preferred ones, and skills ranked higher count more than the rest. One rule does the adding, and nobody re-weights it in the debrief. A weak must-have caps the whole score.

How the score adds up: higher-ranked skills count more; must-haves count more still and a weak one caps the score. Julian Zimmerman, overall 87%: Domain Knowledge 87%, Hard Skills 99%, Core Competencies 75%RANKED SKILLSHigher-ranked skills count moreThe top skill counts most, the next a little less,and so on down the list.Weight by rankMust-haves count more and cap the scoreRatio treats a weak must-have as a key predictorof job performance.Must-have3 of 8 skills
  • The screening round produces a star rating; a rated hiring-manager round produces a percentage
  • The same rule for every candidate on the role, so the scores compare
  • Every rating and its evidence note stay visible, so the debrief is about where ratings disagree

Consistency built in

  • Same four lenses

    Relevance, Approach, Trade-offs, Evidence under every question, in every round.

  • No midpoint

    Four points force a call. There is no safe middle score.

  • Score, then talk

    Ratings go in before the discussion and are kept after it.

  • Humans rate

    Ratio never scores a candidate. It combines what interviewers enter.

Questions teams ask

More questions? Read the full FAQ

More on the product

See the scoring rubric for one of your roles

Bring a job description. In a 30-minute working session we build the interview kit, with a scoring rubric under every question, and you keep it.

Book a Demo