Merit · Prototype

Drop your files — roster, internal ranges, market ranges

Drag in any combination. The file type is auto-detected from columns. Parsing happens locally — files never leave your browser.

Choose file

Recognised columns: roster uses

employee_id, fte, current_salary, annualized_salary, career_level, job_function, job_family, gender, entity, country, salary_eligible

; internal ranges use internal_min/mid/max; market ranges use market_min/mid/max; adjusted gaps use entity, country, career_level, adjusted_gap_mean, adjusted_gap_median (Beqom export shape).

· Roster (0) · Internal ranges (0) · Market ranges (0) · Adjusted gaps (0)

Budget envelope — 4.0% of salary mass

Employees — Avg compa — target 1.00 EUPTD flags — EU Pay Transparency · unadj mean gap > 5% Compression — click to isolate flagged rows

Discretionary merit

€

— / —

—

Non-discretionary

€

— / —

—

Entity Career level Job function —

EUPTD gender pay gap · Cohort grid

Unadjusted mean gap by cohort. Rows flagged when the mean exceeds 5%. Cohorts with fewer than the minimum per-gender sample stay in reading order as “Insufficient data” ghost rows so coverage holes remain visible.

How cohorts are defined, calculated, and flagged

Broad grain default entity × country × career_level — the EU Pay Transparency Directive's “category of workers performing work of equal value” grain. This is how regulators expect the filing.

Fine grain broad + job_function × job_family — drill-down for narrower investigation. The grain toggle above switches between the two.

Disclosure rule Minimum 5 per gender (tunable in the “Disclosure threshold” input — Denmark uses 8 per DISCO group; other member states set their own). Cohorts below the threshold render as greyed ghost rows to preserve the coverage-hole picture in the filing.

Gap math (mean_M − mean_F) / mean_M — signed, where positive values mean women are paid less. Flagged when the result exceeds the 5% materiality threshold. Adjusted gap (when a Beqom-style export is uploaded) is shown alongside in a separate column group to prevent reader-conflation.

Non-binary handling Counted in cohort totals and surfaced in the grid footnote, but excluded from binary F/M means and medians — the directive's reporting scope is binary. NB representation is acknowledged, not erased.

Peer cohort compression A separate per-employee signal: flags anyone sitting > 5% below their fine-grain cohort median. This drives the Compression stat chip in the cockpit and the magenta dots on the Landscape. It is not the EUPTD metric above — don't conflate them.

Cohort grain

Disclosure threshold per gender

The landscape

Compa ratio × Performance. Magenta dots are flagged for pay compression (>5% below fine-grain cohort median). Click any dot to inspect their cohort.

Roster

Click any row for the cohort lens (visible on the Landscape tab). Type a proposed salary and pick a justification to see the budget pools cascade live.

0 edits · €0 uplift · merit €0 · non-disc €0

Name · Gender	Function / Family	CL	Annualised	Proposed	Justification	Compa	Performance	Status

Cohort lens

Select someone to see how they sit against their peers.