Measuring & closing pay gaps by group

The EU Pay Transparency Directive (EUPTD) requires employers to measure and report pay gaps by group, where groups are defined as categories of workers or workers performing work of equal value. Within the PayAnalytics platform, you can easily analyze, understand, and close these pay gaps within each group using the Close pay gaps on a per-group basis setting within your pay equity analysis configuration.

What is a "group"?

Under the EUPTD, organizations must compare the pay of men and women within comparable categories of workers representing groups of workers with the same level of skills, effort, responsibility and working conditions, classified in a non-arbitrary way using gender-neutral and objective criteria.

The key aspects to consider in this context are the following:

• The categories of workers may be dictated by local legislation

• Worker categories might need to be approved by worker representatives

With this in mind, if job categories are not predefined, we recommend a rather broad approach based on grades or role levels. Job family can be considered but it may be risky if not well calibrated (male-dominated or female-dominated job families for instance can be problematic).

In PayAnalytics, these categories are represented by your grouping column (such as Global Grade).

Category definition

Each group will have its own unadjusted and adjusted pay gap, helping you identify where differences exist and how they vary across your organization.

Configuring your analysis

To measure, interpret and close pay gaps by group, you first need to set up your analysis. To do so, proceed as follows:

1. Import your data and review your initial unadjusted pay gap under the Data summary section (under Employee Overview in the main navigation bar) for example 10.8%. (For more information about data import, see Uploading your data. To find the Data summary section, refer to Interpreting pay gaps.)

2. Run your analysis by clicking Run analysis in the top right corner of the Employee Overview page.:

   • Select Gender as your demographic, since under the EUPTD, pay gaps are measured between men and women.

   • Select Total compensation as your pay measure. Typically this is the gross annual pay, including base salary, variable compensation and allowances.

3. Add the selected objective factors that explain pay differences (such as tenure, performance rating or location).

4. In the Set pay equity target section of the analysis configuration page, check the Enable raise suggestions box, and then check the Close pay equity gaps on a per-group basis box.

Enabling the analysis of pay gaps by group

This is a tip message.

When worker categories are defined by a combination of factors, do not use the combined category as a single objective factor. Instead, include each underlying factor separately in the model (for example, grades and job families as distinct variables).

Additionally, the EUPTD Directive states that workers' representatives may challenge the validity of the objective factors.

Defining pay gap targets

You can define a target adjusted pay gap per group. This value represents the goal the system will work toward when computing raise suggestions, knowing that:

• Any group with an unadjusted pay gap above 5% has to be justified by objective factors, leading to the groups adjusted pay gap.

• Many organizations set a 5% threshold, since groups with adjusted gaps above 5% require a joint pay assessment under the EUPTD.

We recommend setting a more ambitious target (for example 0-2%) to promote true pay equity and avoid structural differences that are difficult to explain to underpaid genders.

This is an note/information message.

The adjusted pay gap target applies per group, not to your overall company average.

Pay gap target definition

Choosing a reference group

The reference group is the demographic group (such as men or women) against which pay gaps are calculated, and which determines raise eligibility. To choose the reference group, you can let the system automatically assign it (recommended) or set it manually yourself.

To assign the reference group, the system follows predefined reference group eligibility rules based on headcount to ensure statistical significance.

For example, if one gender has fewer than seven employees in a group (the default threshold), that gender be used as the reference group. For more information about this and other eligibility criteria, refer to Interpreting pay gaps.

This is a tip message.

If needed, you can adjust this threshold size under Settings > System settings, but we recommend that you do so with caution as small group sizes can lead to unreliable reference group comparisons and raise suggestions.

This is an note/information message.

The adjusted pay gap target applies per group, not to your overall company average. For the EUPTD, your pay gaps are by default measured and reported with male as the reference group. Ahead of reporting and particularly within the remediation phase, we recommend letting the system automatically assign the reference group, as it will allow you to address pay gaps in favor of women or men.

Reviewing your results

After running a pay equity analysis based on groups, your next steps are:

1. View pay gaps before and after raises. In the example below, the unadjusted gap decreased from 10.8% to 8.6%, and the adjusted gap (after accounting for objective factors) improved from 4.6% to 0.9%.

Pay gap review

2. You can then go to the Pay gaps by group section and observe your results there:

   • The estimated adjusted pay gap before and after raises is displayed for each group (for example, Grades 1–11).

   • You can quickly identify which groups exceed your defined adjusted pay gap target (example, 5% or 0%) after raises and filter to review them. In the below example, 3 of 12 groups exceed the target:

Pay gap per group visualization

Understand unexpected results

Certain groups might still exceed the defined target after applying raises. This is most of the time due to:

• Reference group eligibility: for instance, the underpaid group is the defined as the reference group and therefore is not eligible for raises. This happens because the overpaid group is proportionally too small to be the reference group.

• Raise eligibility: the underpaid group is too small to be eligible for raises (<7 by default).

This rule prevents the system from suggesting inappropriate raises. For instance, large positive outliers within a small demographic could cause the system to suggest raising many salaries to match their level which can disturb internal equity and the overall gender pay gap.

To analyze and confirm this, we recommend that you perform the following checks:

• Within the columns of the Pay gaps per group section, make sure the reference group headcount and the demographic group headcounts are displayed to view group sizes. Are the demographic groups small compared to the reference group and/or below 7?

• Check the reference group assignment. Is it the underpaid group?

In these cases, we recommend reviewing the affected groups manually (for example, Groups 2, 6, and 7 in the below demo) and assessing adjustments on a case-by-case basis.

Investigation suggestions for unclosed gaps

This is a tip message.

Check out our FAQs to find common reasons why pay gaps within groups may not be closed.