The Six Candidate Evaluation Dimensions, Explained

Most evaluation frameworks ask interviewers to rate candidates on "competencies" like communication, leadership, or culture fit. These labels feel rigorous. They are not. They give different evaluators the freedom to mean entirely different things while producing identical numbers — and that arithmetic consensus disguises disagreement rather than resolving it.

What practitioners actually need is a small set of dimensions that (a) map to the work the role demands, (b) can be scored from observable evidence rather than impressions, and (c) are stable enough to compare candidates in head-to-head evaluation and track an employee's growth over time in promotion decisions. Verdict's six-dimension model was built to satisfy all three.

Why Generic Competencies Fall Short

A competency label names a desired trait without specifying what evidence would place a candidate at one level versus another. "Strong communicator" could mean someone who writes lucid Slack messages, someone who runs all-hands without notes, or someone who convinced a board to fund a pivot. These are meaningfully different. Scoring them on the same label produces noise, not signal.

The deeper problem is attribution. Competency ratings are typically awarded at the end of an interview, when the interviewer's overall impression — shaped by familiarity bias, verbal fluency, and candidate anxiety — has already contaminated the individual dimension. Verdict's methodology avoids this by anchoring every score to a named evidence item: a CV bullet, a verbatim interview phrase, or a portfolio artifact. No evidence, no score.

The Six Dimensions

Each dimension is defined below with the evidence type that drives the score and a worked example drawn from the kind of CV text that hiring teams encounter daily. Scores run 1–5 on each dimension; calibration against the job description is covered in the next section.

1. Capability

Definition: Can the candidate execute the technical or functional work at the level the role requires?

This is the floor. Trajectory and Influence mean nothing if the person cannot do the job. Capability is scoped to the specific requirements of the open role — a 4 in Capability for a junior data analyst role is different evidence from a 4 for a principal machine learning engineer.

Evidence types: Take-home exercises with visible reasoning; portfolio artifacts that show the work, not just the outcome; certifications that required demonstrated skill rather than course attendance; case walkthroughs where the interviewer can probe the decision logic.

Worked example:

CV line: "Built a real-time fraud-detection pipeline in Python (Kafka + LightGBM), reducing false-positive alerts from 12% to 3.1% over six months."

This line earns a 4 in Capability for a senior data engineering role: the candidate demonstrates both system design (Kafka) and model tuning (LightGBM), states a measurable outcome, and implies iterative work over a meaningful timeframe. It would score 3 for a staff principal role where the expectation is system architecture across multiple teams — the pipeline is real, but there is no evidence of cross-team design ownership, trade-off documentation, or other teams adopting the approach.

2. Track Record

Definition: Has the candidate produced measurable outcomes at a scope and scale consistent with the role?

Track Record is the pattern of outputs, not a single achievement. The distinction matters: one lucky win is noise; three wins at increasing scope is signal.

Evidence types: CV bullets with quantified outcomes (revenue, cost, time, error rate, coverage) attributed to a defined scope; progression of impact across roles; outcomes that survived the candidate's departure from the project.

Worked example:

CV line: "Grew enterprise ARR from $1.1M to $4.7M over 24 months by restructuring the renewal process and adding a land-and-expand motion."

This earns a 4 in Track Record for a Head of Sales role. The outcome is quantified, the timeframe is stated, and the candidate names the mechanism (renewal restructure + land-and-expand), which makes it verifiable and distinguishes personal contribution from market tailwind. A 5 would require evidence of structural impact that persisted after the candidate left — for instance, the motion is now the company's standard playbook.

Lower-score anchor (2):

CV line: "Contributed to sales team efforts that helped grow revenue during my tenure."

This earns a 2. There is no quantified outcome, no stated scope, no named mechanism, and the phrasing ("contributed to," "helped grow") makes individual attribution impossible. A Head of Sales candidate whose CV reads this way for their most recent role has not demonstrated that Track Record is even measurable — the evaluator is scoring on faith, not evidence.

3. Trajectory

Definition: Is the candidate's rate of growth accelerating, holding steady, or decelerating relative to time in career and peer benchmarks?

Trajectory matters most in two scenarios: early-career candidates where the track record is thin and you are betting on the slope, and mid-career candidates where a flat line despite tenure is a warning signal.

Evidence types: Scope and title progression across roles with timestamps; interview answers that show compounding learning (candidate can articulate what they know now that they did not know two years ago, and why); the absence of scope regression when switching companies.

Worked example:

CV sequence: IC engineer → tech lead (2 yrs) → engineering manager, 8 reports (2 yrs) → director, 3 teams (18 months).

This earns a 4 in Trajectory for a VP Engineering role. Each transition adds organizational scope within a reasonable timeframe, and the compression from manager to director (18 months) suggests performance recognition rather than job-hopping. A 3 would apply if the title progression existed but each role was lateral in actual scope — same size teams, similar technical domains, no evidence of expanding responsibility.

4. Influence

Definition: Can the candidate move decisions, priorities, and people outside their formal authority?

Influence is the dimension most commonly inflated by interviewer affect. A candidate who speaks confidently about influence often scores high despite providing no concrete example. The scoring rule is strict: only cross-functional or upward influence events where the candidate changed an outcome count.

Evidence types: Specific examples where the candidate altered a decision made by someone outside their reporting line; budget or roadmap changes attributable to the candidate's advocacy; external publications, talks, or community contributions that shifted practice at an industry level.

Worked example:

Interview phrase: "Finance had already approved the vendor contract. I ran a two-week proof of concept with our own team and presented the cost delta — about $380k over three years. The CFO reversed the decision. We built it internally."

This earns a 5 in Influence. The candidate moved a finalized, cross-functional decision by constructing evidence and presenting it upward. The mechanism is named, the stakeholder is named, and the outcome is concrete. Compare with: "I'm good at getting buy-in" — that earns a 1.

5. Domain Edge

Definition: Does the candidate bring knowledge or relationships that are genuinely hard to replicate through hiring or training?

Domain Edge is the dimension that most justifies a premium. It also the most susceptible to false positives: candidates who have been in a vertical a long time often sound expert without holding proprietary knowledge. The test is whether the edge would transfer — whether the candidate's knowledge or network would accelerate outcomes that a well-prepared generalist could not achieve as quickly.

Evidence types: Publications, patents, or open-source contributions that represent genuine intellectual contribution to a field; deep vertical relationships that are not easily replicated (regulatory contacts, key accounts, research collaborators); demonstrated ability to spot and act on asymmetric information within a domain.

Worked example:

CV line: "Co-authored ISO 27001 gap-assessment methodology adopted by three national certification bodies; used as training material in 14 countries."

This earns a 5 in Domain Edge for an information security leadership role. The contribution is externally validated (ISO adoption), has multiplier effect (14 countries of training material), and is attributable to this person specifically. It would be irrelevant for a role where information security compliance is not a core function — Domain Edge must be scoped to what the role actually requires.

Lower-score anchor (3):

CV line: "Eight years of experience in information security, including two years in a regulated financial services environment."

This earns a 3. The tenure in a relevant vertical meets the baseline for a senior security role, but there is no proprietary contribution, no external validation, and no evidence that the candidate's knowledge goes beyond what any experienced practitioner in that environment would accumulate. It satisfies the requirement without exceeding it.

6. Risk Surface

Definition: Are there observable signals that reduce confidence in the other five dimensions?

Risk Surface is not a penalty for career complexity. It is a structured prompt to surface and name uncertainty before a hire, rather than discover it afterward. A high Risk Surface score (4 or 5) means the evaluator must either resolve the uncertainty through additional reference checking or consciously accept it and plan for it in onboarding.

Evidence types: Unexplained gaps in the CV timeline; a pattern of short tenures without a legible narrative; role-scope inconsistencies between title and described responsibilities; references that are enthusiastic about a candidate's personality but silent on outcomes; prior situations where the candidate's stated contribution contradicts publicly available information.

Worked example:

CV pattern: Three roles, each lasting 11–14 months, each described as "left to pursue new challenge." The CV shows no increase in scope across the three roles.

This earns a 4 in Risk Surface. The tenure pattern alone is not disqualifying — early exits happen. What raises the score is the absence of scope growth alongside the exits, which suggests the candidate may be leaving before outcomes are measurable, not because they were recruited upward. The appropriate response is two structured reference calls focused specifically on transition circumstances and on whether the candidate's stated contributions are corroborated.

How to Score Against a Job Description (1–5)

The six dimensions are not equally weighted across all roles. Calibrating scores to a specific JD requires two steps.

Step 1: Identify which dimensions are load-bearing for this role. A chief of staff role is heavily Influence-weighted. A principal engineer IC role is Capability- and Domain-Edge-weighted. A first sales hire is Track-Record- and Domain-Edge-weighted. Identify the two or three dimensions where a weak candidate would clearly fail the role, and treat those as veto dimensions — a 1 or 2 there disqualifies, regardless of scores elsewhere.

Step 2: Anchor the 3 score to the JD minimum. A 3 in any dimension means "meets the stated requirement for this role at this level." A 1–2 means below minimum; 4 means exceeds; 5 means rare, role-defining excellence. If your JD says "managed enterprise accounts above $500k ARR," then a candidate with three $800k accounts earns a 3 in Track Record — not a 4, because they are meeting the bar, not exceeding it. A candidate with a $3M anchor account and documented expansion from $400k earns a 4.

For head-to-head evaluation of a shortlist, compare candidates on the dimensions you identified as load-bearing first. Do not average across all six — a candidate who scores 5/5/5/5/5/1 on a role where Risk Surface is a veto dimension should not outscore a candidate who is 4/4/4/4/4/3. See how Verdict handles head-to-head ranking in the demo.

Weighting by Role Type

Below are practical weighting heuristics for common role archetypes. These are starting points, not formulas — adapt them to the specific JD.

For promotion decisions, Verdict's Promote mode applies the same six dimensions but re-weights toward Trajectory and Influence, since the candidate has already demonstrated Capability in their current role and Domain Edge is assumed.

FAQ

Why six dimensions rather than three or ten? Three is too few to distinguish between meaningfully different candidates who both "meet the bar." Ten is too many to score reliably from a CV and one or two interviews — evaluator fatigue causes the last few dimensions to receive compressed, undifferentiated scores. Six covers the independent failure modes of a hire (can't do the work, poor outcomes, not growing, can't influence, no edge, hidden risk) without overlap.

Can two candidates who score identically on all six dimensions actually be different? Yes — and the evidence notes in each dimension's scorecard entry will show why. Two candidates who both score 4/4/4/4/4/2 may have arrived at those scores through completely different evidence. The dimension score is a compression; the cited evidence is the real information. Verdict surfaces both.

What if a candidate has no track record in a dimension — for instance, a first-time manager has no Influence score? Score it a 1 with the note "insufficient evidence" rather than leaving it blank. A blank creates ambiguity between "not evaluated" and "not demonstrable." Insufficient evidence is a meaningful signal — it shifts weight to other indicators like Trajectory (are they growing fast enough to close the gap?).

How does this model handle protected attributes? All six dimensions are defined in terms of observable, job-relevant evidence. Protected attributes — age, gender, ethnicity, national origin, disability status — do not appear in the scoring criteria and are not proxied by any dimension as defined. Risk Surface, the dimension most susceptible to implicit bias, is explicitly anchored to career-pattern signals rather than personal characteristics. For more on Verdict's approach, see the methodology page.

Where can I see this model applied to real CVs? Run a free sample analysis at the demo — upload any CV and see all six dimensions scored with verbatim evidence in under two minutes.