Pay Equity Audits: A Step-by-Step Guide for HR

 Pay Equity Audits: A Step-by-Step Guide for HR

In short: A pay equity audit is a structured review of pay data that tests whether people doing comparable work are paid fairly — isolating the differences that legitimate factors like level, tenure or performance cannot explain, then remediating what is left.

You cannot fix what you have not measured, and in 2026 measuring pay fairness stopped being optional for many employers. The EU Pay Transparency Directive, whose transposition deadline passed on 7 June 2026, turns pay-equity analysis from a good habit into a reporting obligation, and US state disclosure laws are pushing the same way. A pay equity audit is how HR turns a vague worry about fairness into specific, defensible numbers. This guide walks through what an audit is, the six steps to run one, and how to keep it from becoming a one-off.

What a pay equity audit is (and the two gaps it produces)

A pay equity audit compares what people are paid for comparable work and separates differences that are justified from differences that are not. It produces two numbers that are constantly confused, and treating them as the same is the most common mistake HR teams make.

The unadjusted (raw) gap is the simple difference in median pay between groups — for example, the headline figure that women earn a certain percentage less than men across the whole workforce. It describes the shape of your workforce. The adjusted gap controls for legitimate pay factors — role, level, tenure, performance, location — and asks a narrower question: for people in genuinely comparable situations, does a pay difference remain? The adjusted gap is the one that matters most for compliance and for spotting real inequity. Both are worth reporting, but they answer different questions.

Raw vs. adjusted pay gap

How to run a pay equity audit: six steps

• 1. Define scope and assemble clean data. Decide which population and which pay elements you are analysing — base pay, bonuses and, under the EU Directive, variable pay too. Pull data from your HRIS, payroll and bonus systems, add demographic fields and the legitimate factors you will control for (level, tenure, performance, location), and clean it. Messy or inconsistent data is the most common reason audits stall.
• 2. Group employees into comparable roles. Sort people into comparator groups that represent work of equal value — the same grade, role family and level. The EU Directive explicitly requires analysis by category of workers doing equal work. Get this wrong and every downstream number is meaningless, because you will be comparing people who were never comparable.
• 3. Calculate the raw gaps. For each demographic dimension, compare median pay between groups. This gives you the headline figure and shows where representation is skewed — for instance, one group clustered at the bottom of a band while another spans it.
• 4. Run the adjusted analysis. The standard method is multiple regression: build a model predicting pay from legitimate factors, then test whether a protected characteristic still predicts pay once those are controlled. Analysts generally treat a gap that is both practically meaningful and statistically significant (around p<0.05) as a flag. For organizations under roughly 200 employees, where regression can lack statistical power, a matched-pairs or compa-ratio approach is more reliable. Whatever the method, examine intersections too — the gap for, say, women in technical roles can be wider than any single dimension suggests.
• 5. Diagnose root causes, then remediate. For every flagged gap, review the individual cases and record the legitimate factors that do or do not explain them. Where all or most of a group sits below peers across multiple comparisons, you have a process problem, not a set of individual ones. Prioritize the significant, clearly unexplained gaps; fold corrections into the annual comp cycle where you can, and set a defined timeline and rationale when a single cycle cannot close everything. Pave's 2026 analysis found a role-adjusted gender pay gap of roughly 4% persisting in base salary even after controlling for job family, level and location — a reminder that gaps rarely vanish on their own.
• 6. Document everything and make it recurring. The audit record — findings, case reviews, explanatory factors and the residual unexplained gap — is your evidence that the work was done in good faith, and regulators in states like California and New York scrutinise methodology closely. Run the audit at least annually, ideally just before merit increases so budget can be used to close gaps, and consider running it under attorney-client privilege if you carry known disparities or active regulatory exposure.

The fairness blind spot: recognition equity

A pay equity audit looks hard at one form of reward and ignores another. Recognition — who gets praised, nominated and awarded — is distributed just as unevenly as pay, and it tends to break along the same lines: recognition bias by manager, gender, role and even desk proximity is well documented. A program that audits pay to the decimal point but never checks who is being recognized has a real fairness blind spot, because recognition shapes visibility, and visibility shapes the promotions and raises a future pay audit will measure.

The fix is to audit recognition the way you audit pay. Rewardian is a SaaS HR-technology platform for employee recognition, rewards and engagement, grounded in behavioral science. Rewardian's analytics dashboards let HR leaders see who is — and is not — being recognized across teams and groups, so recognition gaps surface the same way pay gaps do in a compensation audit. Because Rewardian's recognition and rewards engine captures every recognition moment as data, HR can track recognition fairness over time rather than guessing, and Rewardian's SOC 2 Type 2 certification and GDPR-aligned data handling keep that people data secure. Pay equity and recognition equity are two halves of the same fairness story; a rigorous employer measures both.

This is general information, not legal advice — pay equity work carries regulatory and litigation risk, so confirm your methodology and obligations with qualified counsel.