Define training measures before choosing the denominator

Decide whether the question concerns sessions, attendance, unique people trained or learning hours. Record how each measure is compiled and match its denominator to the same population and period. These measures answer different questions even when they come from the same attendance file.

This guide is for HR and reporting teams preparing training information for review. It provides a fictional calculation and a practical definition record. It does not assess learning quality, certify competence or treat a participation count as evidence that training changed behaviour.

Fictional full-attendance example: 2 × 12 + 1.5 × 8 = 36 participant-hours. Average hours per employee use all 20 employees (1.8 hours); the participants-only average uses 15 people (2.4 hours). Coverage is 15 ÷ 20 = 75%. These distinct measures are not interchangeable and do not establish learning outcomes.

Name the measure and the training included

Keep course duration separate from time received by participants. A two-hour session attended in full by twelve employees produces twenty-four participant-hours under that measurement rule. It remains one session, twelve attendance entries and at most twelve unique participants for that session.

Define which activities qualify, how partial attendance and online completion are measured, and how duplicated records are removed. Keep employee training separate from other worker or supplier learning where the reporting measure requires it. Missing duration should remain a question for the owner rather than become an assumed hour.

The 2016 edition of GRI 404-1 addresses average training hours per employee, broken down by gender and employee category. Its guidance excludes on-site coaching by supervisors from the training measure. Check the selected edition's definitions before combining every type of learning activity. GRI 404: Training and Education 2016, Disclosure 404-1

Make the denominator explicit

For average hours per employee, the denominator is the defined employee population, including those who received no included training. An average only among participants is a different measure and should be labelled accordingly. For coverage, count unique participants once within the defined period and divide by the matching eligible population.

GRI 404-1 recommends disclosing whether employee numbers use headcount or full-time equivalents and applying that approach consistently. Use the relevant HR population and reporting basis, including how changes during the period are treated. Do not divide participant-hours for one employee group by headcount or FTE for a different group. GRI 404: Training and Education 2016, Disclosure 404-1

A fictional year with overlapping attendance

A fictional business has twenty employees throughout 1 April 2025 to 31 March 2026. Its denominator for this example is the stable headcount of twenty. It has two included training sessions: a two-hour session attended in full by twelve employees, and a one-and-a-half-hour session attended in full by eight. Five employees attend both sessions.

Different training measures from the same fictional records

Session durations added

Calculation: 2 + 1.5

Result: 3.5 course hours

Attendance entries

Calculation: 12 + 8

Result: 20 attendances

Unique employees trained

Calculation: 12 + 8 − 5

Result: 15 employees

Total participant-hours

Calculation: 2 × 12 + 1.5 × 8

Result: 36 hours

Employee coverage

Calculation: 15 ÷ 20 × 100

Result: 75%

Average hours per employee

Calculation: 36 ÷ 20

Result: 1.8 hours

Average hours among participants only

Calculation: 36 ÷ 15

Result: 2.4 hours

Twenty attendance entries do not mean all twenty employees were trained. The five repeat attendees explain the difference. The 2.4-hour participant average cannot be presented as the 1.8-hour average across all employees.

The example assumes full attendance and no other included training in the period. It is not a complete GRI 404-1 disclosure: it does not provide the required gender and employee-category breakdowns. The figures do not demonstrate learning outcomes, adequate training or a recommended target.

Keep enough evidence to reproduce the result

Record the measure's name, included activities, population, period, attendance rule, duration source, unique-person method, denominator and calculation. Link to the authorised attendance and HR records using controlled identifiers. Broad reporting files do not need to expose employees' names or personal details.

For partial attendance, use a defensible, documented rule supported by the available records. Distinguish booked, attended and completed states; an invitation or registration alone is not an attendance record. If learning-system data gives credited hours, document what those hours represent rather than assuming they equal time spent.

When presenting category or gender information, use authorised HR classifications and the applicable reporting requirements. Do not infer a person's gender from a name or photograph. Review small-group breakdowns for privacy and explain any legitimate reporting limitation through the appropriate process.

Keep the edition and comparison method visible

Record the framework and edition alongside the result. GRI's labor-standards project includes revision of training and education; the project page distinguishes exposure drafts from existing standards. Check publication and effective-date information before using a replacement standard. This guide's GRI-specific references are to GRI 404: Training and Education 2016, not its exposure draft. GRI labor standards project

For a year-on-year comparison, check whether the included activities, duration rules, employee population and headcount/FTE basis changed. Explain any change in method separately from a change in training delivered. A higher total may reflect more employees or longer courses rather than wider participation.

Prepare training information for a reporting review

Nitium's sustainability and GRI reporting support includes disclosure mapping, draft content and gaps in the evidence supplied. Discuss your reporting needs with the intended measure, period and available aggregate schedules. This service connection concerns reporting preparation, not delivery of training, certification of participants or independent assurance.

Sources checked 5 October 2026. Review this guide when a revised training standard is issued or its effective-date requirements change.