Employee data management — columns and working with the table

The header contains four main buttons.

Key rule: any change in the table requires clicking Update database. Without this, the changes are not reflected in the calculation.

The section above the table allows one field to be changed in bulk for selected or filtered rows.

If rows are selected, the change applies only to those rows. If nothing is selected, the change applies to all filtered rows.

Each column header has a filter icon. Clicking it opens a filter and adds a blue chip above the table in the form Column name: value ×. A Clear filters link removes all active filters. When a filter is active, the pagination count changes accordingly, for example from 285 records to 30 filtered records.

Below the table is a standard paginator showing the number of displayed records, numbered pages and Previous / Next buttons.

The Cancel without saving button, shown with a red icon, closes the panel and discards changes. To save changes permanently, the user must explicitly click Update database.

Columns are grouped into three logical blocks: identification and classification columns, remuneration columns and supplementary columns. Each column has a methodological meaning and must be interpreted consistently.

Internal employee identifier. It is used as the key for linking records and for addressing specific employees in detailed views.

The employee’s age. The format depends on Analysis parameters and may be age in years, date of birth or start date.

Employee gender. Accepted formats are flexible according to the client’s preference: 0 / 1, m / f or m / ž.

Length of service in years. The client defines what this value represents. It may mean years in the current position, years in the company or total years of experience. Consistency across the entire dataset is essential.

The highest education level entered as a code defined by the client in Code lists under Analysis parameters. Typical levels are primary school, secondary school, university and doctorate. Each level is mapped to a number of education years, which is then used in the regression.

Binary attribute 0 / 1 indicating whether the role is scarce on the labour market. Value 1 means the employee has a rare skill with limited market availability, which can explain higher remuneration. Value 0 means a commonly available profile.

Performance evaluation from a regular formal review process, such as annual, semi-annual or quarterly evaluation. It can be used as a control variable in the regression.

Free-text column for the client’s internal needs, usually departments or organisational groups. The application does not use this column in the regression.

Position name according to the job description. This is a key column for mapping to the position list in the Work group estimator.

Three parallel columns allow up to three grading mechanisms to be used at the same time. The platform calculates the adjusted pay gap separately for each variant and the user can switch between them in the reports.

If the client has no grading, these columns may remain empty. The main company-level adjusted pay gap can still be calculated, but charts and reports by grade will not be available for the empty grade dimension.

Seniority level within a position. The client defines its own scale in Code lists, for example Junior / Medior / Senior, or any other internal scale.

These two columns distinguish the contract type. For a monthly contract, the activity rate is entered, for example 100% for full-time or 50% for half-time. For an hourly contract, the number of hours for the reference period is entered. Exactly one of the two columns should be filled for each employee, while the other remains zero.

Remuneration components are organised into four expandable groups. The columns are not totals; they are individual pay components. Values are summed into total remuneration for the regression but remain separated for component-level analysis.

Minimum recommended granularity per employee:

If the client has more granular data, it should be split into detailed columns because component-level analysis becomes more accurate.

Binary attribute 0 / 1 with a specific meaning for supplementary pension savings. Value 1 means the employee pays their own contribution to supplementary pension savings and the employer should fill the corresponding employer component in the pension contribution column. Value 0 means the employee does not pay their own contribution.

Common error: if the column is left at zero for everyone although employees actually pay pension contributions, the employer contribution may look like a pure bonus in the variable component. This can distort the result. This column should be checked before running the analysis.

These are actual hours, not the normative value. Normative values are entered separately in Analysis parameters.

Integer attribute used to divide employees into groups for different calculation views. Typical use is to separate top management from other employees:

The analysis can then include all groups or exclude selected groups such as the board. This is useful because extreme values in top positions can significantly distort the average pay gap.

Free-text columns. Notes have a specific role: they store explanations for individual pay differences for specific employees. In an audit or management discussion, this note is the first explanation of why a difference is not discrimination. Additional information 1–5 are free metadata fields.

Action column with a delete icon for removing rows.

The standard data-preparation workflow is to download the template, complete it in Excel or CSV and upload it back through Load data. For direct changes in the application, use filtering to narrow records and bulk edit to change one field. After each set of changes, click Update database, otherwise changes will be lost.

Before running the analysis, check especially the Employee contribution column, completeness of all fields used by the selected regression variant and consistent meaning of Years of service across the dataset.

Internally, the application works with English column names shown in the Raw data section. Key mappings are: