How to conduct a pay equity audit before June 2026: a step-by-step guide

DIRECTIVE 2023/970

UPDATED APRIL 2026

A step-by-step methodology for running a pay equity audit before June 2026. This page covers what data you need, how to calculate gaps by worker category, and what to do when you find one you can't justify.

150+ employees in your company?
You have
days left

until the law takes effect
and the first reports are due

23 / 27
Countries with active progress
as of today,

TL;DR

  • A pay equity audit covers total pay — base salary, bonuses, and allowances — not just base salary. Many audits undercount gaps by starting with the wrong data scope.
  • Worker categories must be built on job evaluation criteria, not job titles. Two people with different titles may be doing work of equal value and must be compared.Without it, you can find gaps but not assess whether they are defensible.
  • Performance data is a required audit input, not a post-audit add-on. Under Article 4.4, performance qualifies as an objective justification factor for pay differences. Without it, you can find gaps but not assess whether they are defensible.
  • A gap above 5% in any worker category isn't automatic non-compliance, but it triggers the justification obligation. Under Article 18, the burden of proof is on the employer — documentation must be retrievable, not just existent.

Every article about the EU Pay Transparency Directive tells you to run a pay equity audit. Almost none of them explain how.

If you're an HR manager or HR director preparing for compliance before June 2026, you've likely read the overview. You know you have gaps to find, obligations to meet, and a deadline to hit. What you need now is a working process: what data to pull, how to structure the analysis, and what to do when the numbers reveal a problem.

This guide covers everything: a five-step audit methodology, a clear picture of where most audits break down, and a checklist you can use during the actual process.

What a pay equity audit is (and isn't)

A pay equity audit is an analysis of your compensation data that identifies whether pay differences between men and women in the same or equivalent roles can be explained by objective, gender-neutral criteria.

Under the directive, that definition carries specific legal weight. The audit is not a one-time exercise you run before the reporting season. It is the analytical foundation for every pay gap report, every employee information request response, and every subsequent joint pay assessment. The quality of your audit determines whether you can answer the question regulators will actually ask — not "do you have a gap?" but "can you explain it?"

One clarification before anything else: "pay" under Article 3(1)(a) of the directive includes all pay components — base salary, bonuses, allowances, and benefits paid directly or indirectly by the employer. Many companies audit base salary alone and report results they cannot defend, because the bonus gap or allowance distribution tells a different story. Your audit scope needs to cover everything.

Before you start: what data you need

Running a pay equity audit requires data from several sources. For most companies, that data lives in at least three separate systems, and pulling it into one place is harder than the analysis itself.

The required inputs are:

  • Employee list with gender data, segmented by role and worker category;
  • Full compensation records: base salary, bonuses, allowances, variable pay, and any benefits with a monetary value;
  • Seniority and progression history: start date, role changes, pay band progression milestones;
  • Performance data: ratings from the last one to two review cycles, OKR or goal completion rates, competency assessments.

Performance data belongs in this list from the start. Under Article 4.4 of the directive, pay structures must be based on objective, gender-neutral criteria — including skills, effort, responsibility, and working conditions, plus any other factors relevant to the specific role. Performance is the most common "other factor" employers will need to document. An audit without performance data can tell you that a gap exists. It cannot tell you whether that gap is defensible. That makes it half an audit, and a potentially misleading one.

For most SMBs, pulling this information means exporting data from HRIS, payroll, and performance tools, reconciled in a spreadsheet. That reconciliation step is where audit accuracy falls apart. Version mismatches, missing records, and inconsistent categorization can invalidate findings before you've calculated a single gap. Performance data is also where errors are least visible: if your performance tool doesn't export cleanly into your compensation file, the justification evidence disappears from the audit before anyone notices.

In Mirro's March 2026 readiness study, none of the HR leaders interviewed had a fully integrated system for compensation, performance, and reporting. The majority were still manually assembling this data. If that describes your setup, you should start building reconciliation time into your audit plan.

Defining your worker categories

Worker categories are the units of analysis for every gap calculation, and this is where most audits go wrong.

Under Article 9(1)(g) of the directive, employers must report the gender pay gap "by categories of workers" — not by job title. Two people with different titles may be doing equivalent work, and the directive requires you to compare them. A junior software engineer and a mid-level developer may or may not belong in the same category. A job evaluation using gender-neutral criteria — skills required, effort involved, level of responsibility, working conditions — determines whether they should be compared.

Worker categories need to be built before the audit, using a documented job evaluation framework, not assigned during the analysis based on whatever groupings are most convenient. Getting categories wrong at this stage flows through every calculation that follows.

*[A full breakdown of gender-neutral job evaluation methodology will be covered in a forthcoming guide on gender-neutral job evaluation and classification.]*

Running the audit: a five-step process

Step 1: Calculate mean and median pay gaps by worker category

For each worker category, calculate:

  • The mean pay gap: average pay for male employees minus average pay for female employees, expressed as a percentage of male average pay - Article 9(1)(a)
  • The median pay gap: the midpoint of the pay distribution for each gender - Article 9(1)(c)

Both are required under the directive. The median is the more reliable number in small worker categories: a single high earner distorts the mean significantly, while the median stays representative. Running only mean calculations can mask a real gap or create the appearance of one that isn't there.

Report these calculations per worker category, not as a company-wide average. A company-wide average of 2% can conceal a 9% gap in one category that triggers mandatory action.

Step 2: Segment by pay components

Once you have base salary gaps by worker category, repeat the calculation for each pay component separately:

  • Bonus pay (mean and median gap)
  • Allowances (travel, housing, mobile, etc.)
  • Variable pay and commission
  • Any other benefit with a quantifiable monetary value

This is where gaps that base salary analysis misses tend to appear. Bonus allocation is frequently less structured than base salary, which means the gap in variable pay is often larger and harder to justify. The directive requires reporting on "the gender pay gap in complementary or variable components" — Article 9(1)(b) — and separately the median gap in those same components — Article 9(1)(d). Running this analysis before the formal report means no surprises.

Step 3: Map results against your pay bands

For each worker category, map where men and women sit across pay quartiles. The directive requires reporting on "the proportion of female and male workers in each quartile pay band" — Article 9(1)(f).

If women are concentrated in the lower two quartiles of a category while men are distributed across all four, that is a structural distribution problem — not just a gap to document but a pattern that signals something systematic in how pay decisions have been made. A gap calculation alone will not show this. The quartile view often reveals the underlying cause of the gap.

Step 4: Assess whether each gap can be justified

For every gap above 5% in any worker category, work through an evidence check against the Article 4.4 criteria. The case for why performance data is a legal requirement in this process is covered in our analysis of performance data and pay transparency compliance. The operational question here is straightforward: given this specific gap in this specific category, what documentation do you have?

We advise you check each of the following:

  • Performance ratings (last one to two cycles): Do higher-paid employees in the category consistently rate higher? Also check whether your rating system is standardized enough to support comparison — subjective or inconsistently applied scales do not satisfy Article 4.3's objective criteria requirement.
  • OKR or goal completion rates: Are pay differences traceable to documented output differences? Completion percentages are concrete evidence. Vague recollections of who "performed better" are not.
  • Performance ratings (last one to two cycles): Do higher-paid employees in the category consistently rate higher? Also check whether your rating system is standardized enough to support comparison — subjective or inconsistently applied scales do not satisfy Article 4.3's objective criteria requirement.
  • Competency assessments: Are there documented skill or qualification differences that correspond to the pay gap? The documentation needs to exist on record, not be reconstructed from memory during a compliance review.
  • Seniority and progression history: Has the higher-paid employee progressed through a structured pay band over time, with documented milestones? Tenure alone is not sufficient — the progression needs to be traceable through a defined structure.

A gap is defensible if it traces to at least one of these criteria with supporting documentation. It is not defensible if the explanation is negotiating history, undocumented convention, or general impressions about contribution.

Under Article 18, the burden of proof shifts to the employer once a worker demonstrates a prima facie case of pay discrimination. Documentation must be retrievable, not just existent.

Step 5: Document findings and flag action items

Before any external reporting, produce an internal audit record that categorizes each gap:

  1. Justified: documented evidence against at least one Article 4.3 criterion
  2. Partially justified: some evidence exists but is incomplete or inconsistently documented
  3. Unjustified: no objective basis found

Category B requires an evidence-building plan before the reporting deadline. Category C requires corrective action. If a gap in category C exceeds 5% and cannot be addressed in time, a joint pay assessment with workers' representatives becomes a legal requirement under Article 10.

What to do when your audit finds an issue

Finding an unjustifiable gap is not a crisis, but it does have a defined response path.

Article 10 establishes the sequence: if a pay gap report reveals a gap of 5% or more in any worker category that cannot be justified with objective criteria, the employer must initiate a joint pay assessment with workers' representatives. The assessment analyzes the causes of the gap and defines remedial actions. Under [Article 10(1)(c)], the joint pay assessment obligation is triggered if the unjustified gap has not been remedied within six months of the pay report submission date.

If remediation involves pay adjustments, those adjustments need to be grounded in the same objective criteria the audit uses for justification. Closing a gap without addressing the underlying structure and without documentation creates a new explainability problem at the next audit cycle.

How often you need to run the audit

Your reporting deadline determines minimum audit frequency, but don't let reporting cycles be your only trigger.

Companies with 250 or more employees report annually from June 2027. For them, the audit needs to run on an annual cycle, with lighter quarterly monitoring to catch emerging gaps before the formal report. Companies with 150 to 249 employees also begin reporting in June 2027, but every three years. Companies with 100 to 149 employees start in June 2031.

Regardless of company size, the transparency measures in the directive (salary ranges in job postings and employee information rights) apply from June 2026, not from your first reporting deadline. The underlying pay structure needs to be auditable from the day the directive goes live. That means this year, not when your first report is due.

Full reporting schedule and deadlines by company size in the EU Pay Transparency Directive deadlines and compliance checklist.

How Mirro supports the audit process

The methodology above assumes you can assemble compensation and performance data and seniority records in one place without data loss. For most HR teams, that assumption is the actual problem.

Pulling audit inputs from three or more separate systems and reconciling them in a spreadsheet takes significant time and introduces risk at every step. Performance records get dropped in the export. Salary figures don't match payroll because the data was pulled on different dates. Worker category assignments in the HRIS don't align with the groupings used in the performance tool.

Mirro removes the reconciliation step. Compensation and performance data (ratings, OKR completion, competency assessments, seniority records) are stored in the same platform. When you run a gap analysis, performance context for each employee is already linked to their salary record. Step 4 from our checklist, the justification check, doesn't require switching to a second system because the evidence you need is in the same view as the gap you're assessing.

Because performance data is captured during normal review cycles rather than imported specifically for audit purposes, the audit can be run quarterly without rebuilding the dataset from scratch each time.

Pay equity audit checklist

Use this checklist to track progress through each stage of the audit. Each item maps to a step in the process above.

1
Phase
Data preparation
 
Employee list compiled with gender data, role, and worker category
 
Compensation data collected: base salary, bonuses, allowances, variable pay, benefits
 
Performance data collected: ratings (last 1–2 cycles), OKR completion rates, competency assessments
 
Seniority and progression history included for each employee
 
Worker categories defined using a job evaluation framework, not job titles
2
Phase
Gap analysis
 
Mean pay gap calculated per worker category Art. 9(1)(a)
 
Median pay gap calculated per worker category Art. 9(1)(c)
 
Mean and median gaps calculated separately for complementary or variable components Art. 9(1)(b) Art. 9(1)(d)
 
Pay quartile distribution mapped per worker category Art. 9(1)(f)
3
Phase
Justification review
 
Each gap above 5% assessed against criteria: performance ratings, OKR completion, competency assessments, seniority Art. 4.4
 
Each gap categorized as justified, partially justified, or unjustified
 
Internal audit record produced before any external reporting
4
Phase
Action planning
 
Evidence-building plan in place for partially justified gaps
 
Remediation plan drafted for unjustified gaps
 
Joint pay assessment initiated where required Art. 10
 
Next audit date scheduled

Running the audit is manageable once the methodology is clear. The harder task, for most organizations, is having the underlying data in a state where the audit produces reliable results—worker categories that reflect actual job evaluation, performance records that are documented and consistent, and compensation records that include every pay component from the start.

Get that data structure in place before June 2026, and the audit itself becomes a reporting exercise rather than an emergency.

Book a demo to see how Mirro supports your pay equity audit.

*This article is part of Mirro's EU Pay Transparency Directive resource center. For a full breakdown of directive requirements and employer obligations, see our guide to EU Pay Transparency Directive requirements.*

FAQ

  • Does every company need to conduct a pay equity audit?
  • The directive's pay reporting obligations apply to companies with 100 or more employees. But the transparency measures — salary ranges in job postings and employee pay information rights — apply to all employers from June 2026. Any company that wants to respond defensibly to individual pay information requests needs the underlying audit infrastructure in place, regardless of size.

  • What's the difference between a pay equity audit and the mandatory pay gap report?
  • The pay gap report is the external output: mean and median pay gap figures by worker category, submitted to authorities on a defined schedule. The pay equity audit is the internal process that produces those figures — and, more importantly, assesses whether each gap is defensible. You need the audit to complete the report accurately. Without it, you're reporting numbers you haven't assessed.

  • What happens if I find a gap I can't justify?
  • An unjustifiable gap above 5% in any worker category triggers a joint pay assessment with workers' representatives under Article 10. The assessment identifies root causes and defines corrective actions. If the gap is not remedied within six months of the pay report submission date, the joint assessment obligation becomes mandatory.

  • What if my performance data isn't documented well enough to use as justification?
  • Then the gap isn't fully justifiable, and that's a compliance risk. Under Article 18 the burden of proof is on the employer. Documentation that was never created cannot be reconstructed for a compliance review. If your performance records are inconsistent or incomplete, the priority before June 2026 is establishing a standardized review process — not running the audit and hoping the gaps are small.

  • How often should I run a pay equity audit?
  • Minimum frequency is tied to your reporting cycle: annually for companies with 250+ employees, every three years for those with 100–249. But the transparency obligations apply from June 2026 regardless of your reporting deadline. An audit run only once every three years leaves you unable to respond defensibly to individual employee information requests in the years between reports. A lighter quarterly monitoring process between formal audits is good practice.