Pay Gap Reporting: 7 Questions Comp & Ben Teams Ask Today

1. How many reporting obligations are there really?

Many companies plan around a single report, while the directive creates several — with different recipients and different operational complexity. It is worth separating three obligations that follow from the law and one that is not required but proves indispensable anyway.

• Aggregate report to the authority (in Poland, to Statistics Poland / GUS) — gap indicators calculated for the whole organisation; technically the simplest, but it requires 100% accurate data.
• Internal report for employees and worker representatives — the same data scope, but unions or the works council may ask for details; a communication challenge, not just a numerical one.
• Individual reports on an employee’s request — an employee may ask how their pay compares with people doing work of equal value. The answer must be prepared within the statutory deadline and use current data.
• Management reporting (optional) — not required by law, but without it the board cannot see how ongoing pay decisions affect the gap.

The most common mistake: a company focuses on the aggregate report and overlooks the individual ones. Yet these are the biggest operational challenge, because they arrive irregularly and require fresh data.

2. What is already certain and what may still change?

Delaying preparation until the final shape of the Polish act is set is a risky strategy — the methodology and the scope of the reported data will not change. The foundation follows directly from the directive and is fixed: the reporting duty (aggregate, internal, individual); the way the gap is calculated (the difference in median or average pay of women and men); a broad definition of pay; mandatory job valuation and four criteria (skills, effort, responsibility, working conditions); the duty to group workers and to explain differences or take remedial action.

In the Polish implementation, several elements are still taking shape: the deadline to respond to an employee’s request (the draft says 30 days, the directive allows 2 months), a possible cap on the number of requests, the level of penalties, how often data must be refreshed, and the extent of worker representatives’ involvement in the valuation methodology. But none of this changes the fact that data must be collected, jobs valued and the gap calculated.

3. Is job valuation mandatory?

Yes. It follows directly from the directive (Art. 4). Without valuation there is no basis for grouping workers, and without grouping the gap cannot be calculated as the law requires. It is also worth noting a distinction: a pay grade in the salary scale is not the same as a group doing work of equal value. A truck driver and an office specialist may sit in the same pay grade yet — given different working conditions and responsibility — belong to different comparison groups.

4. The “gap” is not one number — it is three different concepts

• Raw (unadjusted) gap — the difference in median or average pay of women and men across the whole organisation, without breaking it down by role. This is the indicator reported to GUS — the simplest, but with the least analytical value and the easiest to misinterpret.
• Gap within groups of equal-value work — the correct indicator from the directive’s standpoint; it compares women’s and men’s pay within the same groups. It requires valuation and correct grouping.
• Adjusted gap (regression model / Oaxaca-Blinder decomposition) — an explanatory tool: it shows how much of the difference is attributable to gender-neutral factors (tenure, education, performance, overtime) and how much remains unexplained. The directive does not require it, but for larger populations it becomes practically indispensable for defending the gap.

That third number is the heart of gap management — because it is the unexplained part, not the raw difference, that is the real risk signal. In Levelly.ai, we show the unadjusted and adjusted gap side by side with the Oaxaca-Blinder decomposition, so it is immediately clear how much of the difference structure explains and how much calls for a decision.

5. What if the gap exceeds 5% and cannot be justified?

The directive sets a clear path. If a company can explain the difference with objective, gender-neutral criteria, the matter is closed. If not, it has six months from the report to remedy it. A joint pay assessment — detailed, category by category, conducted with employee representatives — is triggered only when three conditions are met together: the gap in a given category is at least 5%, it has not been justified by objective criteria, and it has not been removed within that six-month period. This is the scenario companies want to avoid — which is why the ability to explain differences in a documented way, ahead of time, matters more than the report result itself.

6. How often to measure the gap — yearly or monthly?

The answer depends on what “current data” will ultimately mean. A yearly aggregate report only requires a file import (Excel or CSV). But the information duty — individual reports on request — requires data current as of the request, and a request may arrive in any month. In practice, this pushes companies towards monthly updates and integration with the HR/payroll system. It is also why the gap is best treated as a process monitored over time rather than a one-off report — which is exactly how monitoring and the AI Insights layer in Levelly.ai work, flagging changes before they become a problem.

7. The hours trap: equal rates, yet the gap still grows

The directive requires converting pay to hourly rates based on normative hours, which change monthly. This is a source of counter-intuitive results. A group at an identical hourly rate can show a large gap simply because one gender regularly works overtime while the other works reduced hours under the parental entitlement for young children or is on long-term leave. No system detects the cause automatically — you have to add the right variable (e.g. average overtime) and check whether it explains the difference.

A separate challenge: the wave of questions after individual reports

This is the least talked-about part, yet it can cost more than the calculation itself. Many employees do not distinguish gross from net pay, do not know the concept of a comparison group, or realise that employer costs beyond salary account for about 20%. Releasing an individual report will trigger questions HR is obliged to answer on time. It pays to prepare in advance: an employee FAQ, training for HRBPs and managers, and a full picture of pay (Total Rewards). The individual employee view in Levelly.ai — showing position within the range and comparisons with the female and male medians, with a downloadable report — exists so that these conversations rest on data, not improvisation.

How Levelly.ai organises the process — platform features

The questions above form a single process: from data, through diagnosis and explanation of differences, to compliance and service to the employee. Levelly.ai guides a company through this process in a single tool.

• Data import and validation — you upload a CSV or Excel file, and the platform immediately detects errors and gaps before anything is calculated.
• Pay gap dashboard — key indicators and pay distribution by gender, job level and location; the gap as mean and median, with or without bonuses.
• Level and group analysis (L1–L5) — a box plot of the distribution, women’s and men’s statistics at a given level, and a list of employees with their deviation from the mean.
• Advanced analytics — unadjusted and adjusted gap (waterfall chart), Oaxaca-Blinder decomposition, a risk-group table (critical / high / medium) and the gap by department.
• Individual employee view — position within the range, comparison with female/male medians, pay history and a one-click downloadable report.
• Compliance view for Directive 2023/970 — a “Compliant / Non-compliant” status, legally required metrics, quartiles, flagging of breaches of the 5% threshold and export of a ready report.
• AI Insights and monitoring over time — an interpretation layer in every view and a report history that shows how pay decisions change the gap.
• Group structure — support for a parent company and its subsidiaries provides a coherent view of pay across the group without losing each entity’s context.

When Excel is enough, and when it isn’t

For a one-off aggregate report and a few charts for the board, Excel is enough. The problem starts with recurrence: monthly merging of pay components from many sources, handling individual reports on request and simulating the impact of raises on the gap are the same steps repeated over and over — several days of an analyst’s work each month in a large organisation. This is the point where a calculator gives way to a system.

Every organisation is at a different stage. It is better to have a plan and follow it step by step than to wait for the final act — because data, valuation and methodology will not change; the only thing that will is how much time is left to prepare.

See where your organisation stands: we’ll calculate your gap on a data sample (unadjusted and adjusted) and walk you through the process from data to a directive-ready report. 15 minutes, no commitment. Book a 15-minute meeting →