What Is a Percentile Score? Meaning, Calculation, and Use in Hiring

TL;DR

• A percentile score shows where a candidate ranks relative to a reference group — not how many questions they got right. Hiring teams that confuse the two reject qualified candidates and advance weaker ones.
• 94% of recruiting professionals now rank skills as the top hiring factor, yet most screening systems still surface raw scores that reveal nothing about relative performance (LinkedIn, 2024).
• The most common scoring error in pre-employment assessment: treating a 70th percentile score as a 70% pass rate. They measure entirely different things.
• Testlify’s assessment library spans 3,000+ tests across 4,500+ job roles, each with role-specific norm groups that make percentile scores meaningful by function, not just by population.
• Hiring teams that set role-calibrated cut scores see up to 55% reduction in time-to-hire without sacrificing quality.
• Applying a single fixed percentile cut across all roles inflates rejection rates for entry-level positions and under-screens for senior ones simultaneously.
• The Testlify Percentile Benchmark Framework gives talent teams a three-step repeatable process to build, validate, and recalibrate percentile benchmarks for every role they hire.

What is a percentile score?

A percentile score is a number between 1 and 99 that shows the percentage of people in a reference group who scored at or below a given result. A candidate who scores at the 80th percentile performed better than 80% of the norm group on that assessment, regardless of how many questions they answered correctly.

Percentile scores appear in pre-employment testing, cognitive ability tests, psychometric evaluations, and standardized benchmarks. They are the standard output for any assessment designed to compare candidates against a defined population rather than a fixed pass/fail threshold.

Key Takeaway: A percentile score measures relative performance, not absolute accuracy. Two candidates can score 70% on the same test and land at completely different percentiles if their reference groups differ.

How do you calculate a percentile score?

The standard formula is:

Percentile = (Number of values below the score / Total number of values) x 100

For a 12-candidate cohort where a candidate ranks 9th from the bottom:

Percentile = (9 / 12) x 100 = 75th percentile

That candidate performed better than 75% of the group. The raw score — whether 42 out of 60 or 51 out of 60 — does not appear in the percentile output.

The calculation scales identically whether the cohort has 12 candidates or 12,000. What changes is precision: larger norm groups produce more stable, reliable percentile distributions.

Pro Tip: Always confirm which norm group a percentile score is calibrated against before using it in a hiring decision. A score calibrated against “general population adults” means something very different from one calibrated against “software engineers with 3 to 5 years of experience.”

What is the difference between a percentile score, raw score, and percentile rank?

These three terms appear interchangeably in hiring conversations, but they measure different things:

Raw scores are useful for checking whether a candidate cleared a minimum threshold. Percentile scores are useful for ranking candidates against each other and against a validated norm group. Hiring decisions that rely solely on raw scores miss the relative performance signal that percentile data provides.

Key Takeaway: Use raw scores to set minimum pass gates. Use percentile scores to rank and compare candidates within a qualified pool.

What do percentile ranges mean in practice?

Percentile ranges translate a single number into a performance band that hiring teams can act on:

These bands shift by role. A 65th percentile on a verbal reasoning test may clear the bar for a customer service coordinator but fall below the cut for a senior analyst. The Testlify Percentile Benchmark Framework — covered later in this post — provides a method for setting role-calibrated cut scores rather than applying population-level bands uniformly.

How do percentile scores work in pre-employment testing?

In pre-employment testing, a percentile score is generated by comparing a candidate’s raw score against a norm group: a reference population of people who have previously completed the same assessment. The quality of the norm group determines the quality of the percentile signal.

A norm group can be:

• General population (all adults who have taken the test)
• Job-function specific (financial analysts, software engineers)
• Industry specific (financial services, healthcare, logistics)
• Seniority specific (individual contributors vs. managers)

Testlify’s library of 3,000+ assessments across 4,500+ job roles includes role-specific norm groups built from validated candidate data. A candidate’s percentile score on a financial modeling test is benchmarked against other finance professionals, not a general adult population. That distinction changes hiring decisions.

According to LinkedIn’s 2024 Future of Recruiting report, 94% of recruiting professionals now rank skills as the most important factor in hiring. Generic norm groups undermine skills-based hiring by comparing candidates against the wrong reference pool.

For a deeper look at how assessment data integrates with skills-based hiring, see Testlify’s guide to building a skills-first screening process.

Pro Tip: When evaluating an assessment vendor, ask which norm group is used for each test. If the answer is “general population,” request access to function-specific norms before using the tool for technical or specialized roles.

How do hiring teams use percentile scores to shortlist candidates?

Three applications account for most of the value hiring teams extract from percentile scores.

Shortlist candidates faster

Rather than reviewing every application manually, hiring teams can set a percentile cut score — say, the 70th percentile on a role-relevant cognitive test — and automatically advance candidates who clear it. SHRM data puts the average cost per hire at $4,700, and reducing manual screening time directly reduces that number. Testlify clients using percentile-based shortlisting report up to a 55% reduction in time-to-hire compared to CV-only screening.

Spot skill gaps early

Percentile scores across multiple assessments create a skill profile. A candidate who scores at the 85th percentile on technical skills but the 30th percentile on communication creates a different hiring conversation than one with balanced scores across both. These gaps surface before the first interview, not after a 90-day performance review.

Gallup research shows that poor hiring decisions cost organizations between one-half and two times the employee’s annual salary. Identifying skill gaps at the assessment stage is one of the highest-leverage interventions a talent team can make.

Benchmark current employees

Percentile scores from pre-employment assessments can anchor an internal benchmark. When hiring managers see that top performers in a role consistently score at or above the 75th percentile on a specific competency, that benchmark becomes the evidence-based cut score for future hiring cycles — not a number someone picked in a planning meeting.

See how Testlify’s cognitive ability tests generate percentile data that maps to role performance benchmarks.

What mistakes do hiring teams make when reading percentile scores?

Five errors appear repeatedly in organizations that use assessment scores without formal scoring protocols.

Confusing percentile with percentage

A score at the 70th percentile is not a 70% pass rate. It means the candidate outperformed 70% of the norm group. A candidate who scores at the 70th percentile on a difficult test may have answered fewer than 60% of questions correctly. Treating percentile as a percentage inflates rejection rates and eliminates qualified candidates.

Using a single cut score across all roles

Setting the 75th percentile as the universal hiring bar regardless of role complexity creates two problems at once: it over-screens for entry-level roles where the 50th percentile is a sufficient bar, and it under-screens for senior roles where the 80th or 85th percentile is the evidence-based threshold.

Ignoring norm group relevance

A software engineer at the 60th percentile of the general population may rank at the 40th percentile among engineers with comparable experience — a 20-point difference that changes whether that candidate advances or is rejected.

Treating percentile scores as the only signal

Percentile scores quantify cognitive and skills performance, but they do not capture motivation, culture alignment, or contextual judgment. Testlify’s talent assessment approach combines percentile scores with structured interview data and work sample outputs to build a composite candidate profile that no single score can produce.

Over-relying on a stale benchmark

A percentile benchmark built from a 2021 hiring cohort may not reflect the current candidate pool, role requirements, or skills baseline for that function. Benchmarks should be reviewed and recalibrated at minimum annually, and after any significant change to role scope or required competencies.

Key Takeaway: Percentile scores are most reliable when the norm group is role-relevant, the cut score is evidence-based, the benchmark is current, and the score is one input in a multi-signal evaluation — not the deciding factor on its own.

How do you build a fair, data-driven percentile benchmark for your roles?

The Testlify Percentile Benchmark Framework gives hiring teams a repeatable three-step process.

Step 1: Define the norm group

Select the norm group that matches the role, not the organization. For a data analyst role, use a data and analytics norm group rather than a cross-functional general population. Testlify’s platform surfaces available norm groups by function and seniority at test setup, so talent teams can match before candidates start testing, not after results come in.

Step 2: Set a role-level cut score

Use 90-day performance data from existing employees to identify the percentile range where high performers cluster. If top performers in a role consistently score between the 72nd and 88th percentile on a given assessment, the evidence-based cut score sits near the 70th percentile — not at an arbitrary 75th set by policy.

Testlify’s 100+ ATS integrations allow teams to pull post-hire performance data back into the platform and calculate these correlations directly, without manual data exports or spreadsheet matching.

Step 3: Validate over a 90-day hire cohort

After setting the cut score, track 30, 60, and 90-day performance outcomes for every hire made above and below the threshold. If candidates who cleared the 70th percentile consistently outperform those who did not, the benchmark is validated. If the correlation is weak, adjust the cut score or add a second assessment that better predicts performance for that specific role.

Pro Tip: Do not set a percentile benchmark without involving the hiring manager who owns the role. A benchmark calibrated to the wrong success definition produces a valid score and still a bad hire. Define what “high performance at 90 days” means before the first data point enters the model.

A properly built percentile benchmark, reviewed annually and tied to actual performance outcomes, is one of the few hiring tools that compounds over time. The more hiring data enters the system, the more precise the benchmark becomes.

What does a percentile score of 85 mean?

A percentile score of 85 means the candidate performed better than 85% of the people in the norm group who took the same assessment. It does not mean the candidate answered 85% of questions correctly.

Is a higher percentile score always better in hiring?

For most pre-employment assessments, yes — a higher percentile indicates stronger relative performance. The exception is role fit: an exceptionally high score on a low-complexity assessment may indicate the candidate is overqualified, which carries retention risk for that specific role.

How is a percentile score different from a z-score?

A z-score is a raw statistical value that measures how many standard deviations a result sits above or below the mean. A percentile score is derived from the z-score and expressed as a ranking from 1 to 99. Most hiring platforms display percentile scores because they are easier for non-technical users to interpret and act on.

What is a good percentile cut score for hiring?

There is no universal answer. The right cut score depends on the role, the norm group, and historical performance data from employees in that function. Most organizations find that the 65th to 75th percentile range is a practical starting point for professional roles, with adjustment after 90 days of performance tracking.

Can percentile scores be biased?

Any assessment that uses a norm group built from a non-representative population carries the risk of systematic bias. Hiring teams should audit norm group demographics and validate that percentile distributions are consistent across gender, age, and background before deploying at scale. Testlify provides adverse impact reports to support bias audits before a benchmark is applied to live hiring.

Testlify generates role-specific percentile norms across 4,500+ job roles. Book a demo to see how the Percentile Benchmark Framework works inside your hiring workflow.