A cut score can determine who progresses to interview, who is screened out and whether a recruitment process can be defended when challenged. Yet many employers set assessment cut scores by choosing a round number, copying a previous campaign or selecting the top percentage of applicants. Those shortcuts may feel efficient, but they can exclude capable people and weaken the quality of the shortlist.
To set assessment cut scores well, employers need to connect assessment performance to the actual requirements of the role. The right threshold is not simply the highest possible score. It is the score that separates candidates who are likely to meet a genuine job requirement from those who are unlikely to do so, using evidence that is relevant, consistent and practical.
What an assessment cut score is designed to do
A cut score is the minimum result a candidate must achieve to move forward in a selection process. It may apply to a cognitive ability test, job knowledge assessment, skills test, safety screen or another measure used during recruitment.
For example, an employer recruiting customer service staff may require candidates to demonstrate a minimum level of written communication, numerical accuracy and situational judgement. A technical role may require a stronger threshold on a specific software skill or safety-critical knowledge test. The threshold should reflect what is needed to perform the role safely and effectively, not an arbitrary preference for high test scores.
A cut score is also different from a ranking rule. A cut score answers, “Has this candidate met the minimum standard?” Ranking answers, “How does this candidate compare with other applicants?” Many recruitment processes need both. A minimum threshold can remove candidates who do not meet essential requirements, while ranking helps employers prioritise those with the strongest overall evidence.
Why arbitrary thresholds create hiring risk
A score of 70 per cent may look familiar from school or training contexts, but it has no automatic meaning in a hiring assessment. Difficulty varies between tests, as does the relationship between a score and job performance. A 70 per cent result on one assessment may indicate strong capability. On another, it may be unnecessarily demanding or too low for a safety-critical role.
An overly high cut score can shrink a viable talent pool, particularly in a tight labour market. It may also create unintended adverse impact if some groups are disproportionately screened out. A threshold that is too low has a different cost: recruiters spend time interviewing applicants who do not have the required skills, and hiring managers receive a less useful shortlist.
The commercial objective is not to create the smallest possible shortlist. It is to create a shortlist with enough qualified candidates to support a sound decision, without adding avoidable time and cost to the process.
How to set assessment cut scores for a role
Start with the job, not the assessment
Begin with a clear job analysis. Identify the knowledge, skills, abilities and behavioural requirements that are essential from day one, as well as those that can reasonably be learned after appointment. Speak with experienced job holders, hiring managers and relevant subject matter experts. Review position descriptions, performance expectations, training requirements and any safety or regulatory obligations.
This step prevents a common error: treating every measured attribute as equally important. For an entry-level role, a high level of advanced technical knowledge may not be essential if structured training is available. For a role involving medication, financial approvals or critical infrastructure, the minimum standard for accuracy or safety knowledge may need to be much higher.
Use evidence from the assessment itself
A validated assessment should provide information about its scoring, reliability, comparison groups and intended use. These details help employers understand what a result represents and whether a proposed threshold is reasonable.
Where historical recruitment data is available, examine the relationship between assessment scores and relevant job outcomes. These may include training completion, probation outcomes, quality measures, sales performance, safety incidents or supervisor ratings. The aim is not to chase a perfect correlation. It is to establish whether people who meet a proposed threshold are more likely to succeed in the role than those who do not.
For a new role or a new assessment, local performance data may not yet exist. In that case, use job analysis, expert judgement and assessment validation evidence to set an initial threshold, then review it as hiring outcomes accumulate.
Choose the right type of threshold
Different selection contexts call for different approaches. A criterion-referenced cut score is appropriate when candidates must demonstrate a defined level of competence, such as literacy, numeracy, job knowledge or safety awareness. The score is based on what a person needs to know or do, rather than how they compare with other applicants.
A norm-referenced approach compares candidates with a relevant population. This can help where the employer needs to identify stronger performers from a large applicant pool, such as a graduate program. It should not be mistaken for a minimum competence standard, however. Selecting the top 20 per cent of candidates may be operationally useful, but it does not prove that the remaining 80 per cent lack the capability to perform.
Some roles need a hurdle approach. Candidates must meet a minimum standard on one essential measure before their broader results are considered. This is often appropriate where a capability is non-negotiable, such as a licence-related knowledge requirement or a core technical skill. For most roles, avoid creating too many hurdles. Several strict thresholds can compound exclusion and make the process harder to justify.
Account for score precision and borderline results
No assessment score is perfectly precise. A candidate who narrowly misses a cut score may have a level of capability very similar to someone who narrowly passes. This matters most when the consequence is an automatic rejection.
A practical solution is to create a review band around the threshold. Candidates clearly above the cut score progress. Candidates clearly below do not. Those close to the line can be considered alongside other relevant evidence, such as structured interview results, work samples or verified experience.
This does not mean abandoning consistency. It means recognising the limits of measurement and applying a pre-defined rule rather than making ad hoc exceptions after results are known.
Test the impact before making the score final
Before launching a high-volume campaign, model what the proposed cut score would do to a representative applicant group where possible. How many candidates would progress? Is that number workable for the recruitment team and hiring managers? Does the pass rate differ substantially across relevant groups? Are high-quality candidates being excluded because the threshold is set beyond the actual requirements of the role?
A pilot can be particularly valuable when introducing a new assessment or changing a long-standing process. Run the assessment alongside existing selection methods for an initial period, then compare results with interview outcomes and early performance indicators. This gives the organisation a factual basis for refining the threshold rather than relying on assumptions.
Combine assessment data with structured selection methods
Most hiring decisions should not rest on a single score. Assessments are strongest when they are part of a structured process that may include a consistent interview, work sample, reference checking and appropriate pre-employment screening.
The weight given to each component should reflect the role. A skills test may deserve significant weight for a data-entry position where accuracy is central. For a leadership appointment, cognitive ability, behavioural assessment, structured interviewing and career evidence may each add useful information. The key is to define the decision rules before candidates are reviewed.
RightPeople supports this process by helping employers interpret validated assessment data in the context of the role, rather than treating reports as a simple pass or fail exercise.
Document the decision and review it over time
A defensible cut score has a clear rationale. Record the role requirements, assessment selected, evidence considered, threshold chosen, any review band and the rules for combining results with other selection information. This supports consistency across recruiters and gives hiring managers confidence in the process.
Review cut scores when the role changes, the labour market shifts, a new assessment version is introduced or performance data suggests the threshold is not working as intended. A cut score is not a permanent setting. It is a hiring control that should be monitored like any other part of a quality recruitment process.
The best threshold is one that reflects real job demands, treats candidates consistently and leaves employers with a shortlist they can act on confidently.
