← Back to studio

📈 HR analytics guide

The HR Analytics Framework: 5 Steps to Board-Ready Metrics

Most HR reports drown a board in numbers and starve it of decisions. This is the five-step framework we built HR Metrics Studio around — how to choose HR metrics, calculate them the same way every quarter, benchmark them honestly, and turn them into a story leadership can act on.

1. Start with the decision, not the data

Every metric should answer a question a leader is already asking. If nobody would act differently based on the number, it does not belong in the report. Write the question first — "Are we losing our best people faster than last year?" — then choose the metric that answers it (voluntary turnover of high performers, not headline turnover).

2. Fix the formula before you fix the number

Most HR reporting arguments are definition arguments. Agree in writing what counts as a hire, a leaver, an FTE, and a reporting period — then keep that definition for at least four quarters. Cost per hire, for example, must include internal recruiter cost and advertising, not just agency invoices, or your figure will look artificially strong.

3. Benchmark honestly, in three bands

A number without a reference point is noise. Score each metric against an external benchmark for your industry and a rolling internal baseline, and label it plainly: 🟢 Excellent, 🟡 Healthy, 🔴 Needs attention. Three bands force a judgement; ten shades of grey let everyone avoid one.

4. Write the insight, not the reading

"Turnover is 18%" is a reading. "Turnover rose to 18%, driven almost entirely by operations staff in their first year, which cost roughly the equivalent of nine full salaries to replace" is an insight. Every metric in a board pack should carry one sentence of cause and one sentence of cost or risk.

5. Report at board altitude

Boards want fewer than twenty numbers. Lead with the exceptions, show direction of travel rather than raw tables, keep charts to one message each, and close with the two or three decisions you need from them. Everything else belongs in an appendix.

🗂️ The ten HR metric categories worth tracking

A complete HR scorecard spreads across ten families. You do not need all of them in month one — but a gap in any of them is a blind spot somebody will eventually ask about.

  • Recruitment

    Cost per hire, time to fill, time to hire, offer acceptance rate, quality of hire

  • Payroll & cost

    Labour cost ratio, payroll cost per employee, overtime rate, compa-ratio

  • Retention

    Voluntary and involuntary turnover, retention rate, first-year attrition, regretted loss

  • Diversity

    Gender and representation ratios, pay gap, promotion parity

  • Learning

    Training spend per employee, training hours, completion rate, skills coverage

  • Engagement

    eNPS, engagement score, participation rate, manager effectiveness

  • Productivity

    Revenue per employee, profit per employee, HR-to-staff ratio

  • Absence

    Absence rate, Bradford factor, lost days per employee

  • Safety (ISO 45001)

    TRIR, LTIFR, severity rate, near-miss reporting, audit score, training coverage

  • Performance

    Performance distribution, goal completion, internal mobility, succession cover

❓ Common questions

What is an HR analytics framework?
A repeatable process for choosing HR metrics, calculating them consistently, comparing them to benchmarks, interpreting what they mean for the business, and reporting them to leadership. The framework matters more than any single metric, because it is what makes this quarter comparable to the last one.
Which HR metrics does a board actually care about?
Cost and risk. Voluntary turnover, cost per hire, time to fill, labour cost as a share of revenue, revenue per employee, absence rate, and safety indicators such as TRIR under ISO 45001. Everything else supports those.
How often should HR metrics be reported?
Operational metrics monthly, strategic and board metrics quarterly, with a full annual review covering trend lines and benchmark movement.