How to Calculate Every EU Pay Transparency Directive Reporting Metric — and What it Means for Global Orgs

A lot of companies are preparing for the EU Pay Transparency Directive's reporting requirements by figuring out their gender pay gap figure. That's a start. But the actual reporting obligation under Article 9 of the Directive is considerably more detailed than a single figure, and most companies underestimate what it takes to produce it correctly.

This post breaks down each required metric, how to calculate it, what data it depends on, and what the reporting picture looks like when your organization operates across multiple EU countries.

What Article 9 of the EU Pay Transparency Directive actually requires

Companies with 250 or more employees must report annually, starting with data from 2026 (first submission in 2027). Companies with 150 to 249 employees report every three years. The required outputs are:

1. The mean gender pay gap
2. The mean gender pay gap in complementary and variable components of pay
3. The median gender pay gap
4. The median gender pay gap in complementary and variable components of pay
5. The proportion of female and male workers receiving complementary or variable components
6. The proportion of female and male workers in each quartile pay band
7. The gender pay gap between workers by categories of workers, broken down by base wage and variable components separately

Seven outputs. Each one requires different data, different methodology, and a different layer of infrastructure underneath it.

How to calculate each metric

1. Mean gender pay gap (base pay)

Add up total base pay for all male workers and divide by the number of male workers to get the mean. Do the same for female workers. The gap is expressed as: (mean male base pay minus mean female base pay) divided by mean male base pay, multiplied by 100.

A positive number means men earn more on average. What you need: base pay for every worker, with gender and employment status accurately recorded.

2. Mean gender pay gap in complementary and variable components

The same calculation as above, but applied only to variable and complementary pay: bonuses, overtime, shift premiums, travel allowances, housing allowances, sick pay, and variable pay. This is reported separately from base pay, not combined with it.

What you need: every variable and complementary component tracked and attributed at the individual worker level, broken out by type. A payroll total by worker is not sufficient. The data needs to be disaggregated.

3. Median gender pay gap (base pay)

Sort all workers by base pay from lowest to highest. Find the middle value for men and the middle value for women separately. The gap is: (median male minus median female) divided by median male, multiplied by 100.

The median is less influenced by outliers than the mean. A very highly paid executive can skew mean figures significantly without affecting the median much. The Directive requires both, because they reveal different things about your pay distribution.

What you need: the same base pay and gender data as metric 1, but structured in a way that can be sorted and ranked.

4. Median gender pay gap in complementary and variable components

The same median methodology applied to variable and complementary pay only.

What you need: the same disaggregated variable pay data as metric 2, sortable by worker.

5. Proportion receiving complementary and variable components

For each gender: how many workers in that group received any variable or complementary pay during the reporting period, expressed as a percentage of all workers in that gender group. For example, 80% of men received a bonus versus 61% of women.

This metric is specifically designed to surface structural bias in bonus eligibility, not just bonus size. A company where women are systematically excluded from variable pay programs will show a gap here even if the average bonus amounts look similar.

What you need: a record per worker of whether they received any variable pay during the period, not just the amount.

6. Proportion of each gender in each quartile pay band

Sort the entire workforce by total pay from lowest to highest and divide into four equal groups. For each quartile, calculate what percentage of workers are female and what percentage are male.

A heavily male-skewed upper quartile and female-skewed lower quartile is a structural pattern, not a statistical coincidence. This metric makes that visible.

What you need: total remuneration per worker to rank the population, complete gender data, and a consistent job architecture to define which workers belong in the analysis. If you include contractors in some countries but not others, the quartile distribution becomes incomparable across entities.

7. Gender pay gap by category of worker

For each defined category of workers, calculate the pay gap between men and women in base pay, and calculate it again for variable components. The two are reported separately.

"Category of worker" must be defined using objective, gender-neutral criteria based on the same work or work of equal value. This is the metric that makes job architecture non-negotiable. Without a structured, documented framework for grouping job profiles into comparable categories, you cannot produce this output in a way that is defensible or repeatable.

What you need: a job architecture that maps every job profile to a family and level, applied consistently across the organization, with documentation of the criteria used to define equivalence.

What this looks like for a global organization

Here's where it gets significantly more complex for companies operating across multiple EU countries. Take an organization with 200 employees in Spain, 150 in Germany, and 250 in Poland, each operating as a separate legal entity.

Reporting thresholds apply per entity

The Polish entity, with 250 workers, falls in the annual reporting bracket and must submit its first report in 2027. The Spanish entity at 200 workers and the German entity at 150 workers both fall in the 150 to 249 bracket, which requires reporting every three years. Their first required reports may not come until 2031, depending on the member state's transposition timeline.

Three reports, three national authorities

Each entity submits its report to the relevant national authority in its own country. Each member state can specify its own submission format, filing portal, and deadline within the calendar year. You are not filing one report. You are managing three separate compliance tracks.

"Work of equal value" comparisons stay within each entity

A Senior Software Engineer in Krakow is not compared to a Senior Software Engineer in Berlin for statutory reporting purposes. The analysis is contained within each legal entity. This matters for data structure: your job architecture needs to be consistent across all three countries so that the category definitions are defensible within each entity, even though the reports are separate.

Currency and normalization

Spain and Germany report in euros. Poland reports in Polish zloty. For each national report, figures are in local currency, which is straightforward. The challenge comes when your People Analytics team wants to run a group-level view across all three entities. The Directive doesn't require that, but leadership usually does. Building the normalized view requires a shared data layer that doesn't exist in most multi-country setups.

Data systems in practice

In the scenario above, there is a meaningful probability that the Spanish, German, and Polish entities each use different payroll providers, different HR configurations, and different compensation tracking processes. Producing three consistent, auditable reports from three different data sources requires either centralizing that data before running the analysis, or building a reporting layer that ingests and normalizes inputs from all three. Most companies are currently doing neither. They are doing manual exports and reconciliation, which produces results that take weeks and carry real risk of error.

What the foundation needs to look like

To produce these seven metrics accurately, repeatably, and with confidence, you need four things in place:

1. One HR record per worker with complete and current information on job profile, level, gender, and employment status
2. All remuneration components tracked and attributed at the individual worker level, not just the payroll total
3. A job architecture that groups workers into defensible categories and is applied consistently across every entity
4. A connected system where HR records and payroll data live together, so the report runs from live data rather than a reconciled export.

The report itself is not the hard part. The foundation is.

How Deel helps

Deel approaches pay transparency as a partner across two layers: the software infrastructure that makes reporting possible, and the consulting expertise that helps you build what goes into it.

On the software side, Deel connects job architecture, compensation bands, structured pay cycles, pay gap reporting, and global payroll in one platform across 150+ countries. Job families, levels, and job profiles live in Deel HR's HRIS and feed directly into compensation bands in the Compensation module. Pay gap reporting draws from that same connected data, so the analysis runs on a single source of truth rather than a reconciled export from four separate systems. For organizations operating across any combination of EU member states, each entity's data stays structured, attributed, and ready to report without rebuilding the process each cycle.

On the consulting side, Deel also offers specialist services for organizations that need expert guidance to build the foundation the software depends on. That includes pay equity audits, job architecture design, salary grid and band-setting, manager and HR training, and HR data and reporting. These services are delivered by Deel's in-house consulting team and are scoped separately from the software implementation.

We don't promise guaranteed compliance. What we offer is the infrastructure and the expertise to help your organization set itself up to meet the requirements of the Directive, and to manage those requirements as they evolve across every jurisdiction you operate in.

Book a consultation with a Deel expert.