The characteristic failure of HR board reporting is not a shortage of metrics. It is thirty of them, arranged in a dashboard, with no indication of which three matter this quarter. A board reads it once, asks a question the dashboard cannot answer, and thereafter treats the people section as a formality to be moved through before the numbers that carry consequences.
The fix is subtraction. A board pack is not a management dashboard and should not contain management's operating metrics. It exists to let a group of people who spend a few hours a quarter on your company form a judgement about three things — and to surface the exceptions that require a decision.
The three questions
Every metric that earns a place on the page answers one of these. If a metric answers none of them, it belongs in the appendix or in the management pack.
| Question | What the board is really assessing |
|---|---|
| Can we staff the plan? | Whether the operating plan the board approved is deliverable with the people you will actually have |
| Are we losing people we cannot afford to lose? | Concentration of risk in individuals, and whether departures are the ones you would have chosen |
| Is people cost under control and compliant? | The largest cost line in most mid-market companies, and the one with statutory consequences attached |
The nine
| Metric | Answers | The trap |
|---|---|---|
| Headcount versus plan | Staffing | Reporting headcount alone. The number is meaningless without the plan it is being measured against, and the variance is the story |
| Voluntary attrition, rolling 12 months | Retention | Quarterly figures annualised. Small denominators produce wild swings and invite conclusions the data cannot support |
| Regretted attrition | Retention | Not separating it from total. A model that prices all departures identically implies the optimal rate is zero, which is false |
| Median time to fill | Staffing | Using the mean. One nine-month executive search distorts an average and tells you nothing about the typical requisition |
| Offer acceptance rate | Staffing | Ignoring it. It is the earliest available signal that your compensation position or candidate experience has slipped |
| People cost as a share of revenue | Cost | Comparing to an external benchmark. There is no credible mid-market benchmark; your own trend is the useful comparison |
| Payroll corrections per run | Cost and control | Counting only errors that reached employees. Count every correction, including those caught pre-disbursement |
| Statutory filings on time | Compliance | Reporting it as a percentage when the only acceptable answer is 100%. Report exceptions by name instead |
| Critical roles with an identified successor | Retention risk | Defining critical too broadly. If a third of the company is critical, the metric has stopped meaning anything |
Nine is a ceiling, not a quota. If three of them are flat and uninteresting this quarter, report them in a single line and spend the page on the ones that moved.
What the page should look like
What to leave out
Engagement scores with no action attached. A number that moves two points with no stated cause and no consequence trains a board to discount your entire section. Report engagement when you are asking for something on the basis of it.
Training hours and activity counts. Hours delivered, sessions run, applications received, interviews conducted. These are measures of HR's activity, not of the company's condition. They belong in a functional review.
Headcount without plan variance. Covered above, and worth repeating because it is the single most common inclusion.
Anything you cannot define in one sentence. If the definition needs a paragraph, the discussion will be about the definition.
Benchmarks you cannot source. Quoting an industry average that a director then cannot find is a credibility cost that outlasts the meeting.
Define it before you report it
Most disputes about HR metrics in board meetings are definitional, and they are avoidable. Publish your definitions once, in an appendix, and do not change them mid-year without saying so.
| Metric | Decisions you must make explicit |
|---|---|
| Attrition | Denominator: opening headcount, closing, or average? Does it include probation exits, fixed-term expiries, contractors? |
| Regretted | Who decides, and when? A manager's judgement at exit, or a defined performance and criticality rule? |
| Time to fill | Clock starts at requisition approval or at first advertisement? Stops at offer acceptance or at start date? In India the gap between those two is often 90 days |
| People cost | CTC, or CTC plus recruiting, benefits, and employer statutory contributions? Contractors in or out? |
| Critical role | Criticality by revenue exposure, by replacement difficulty, or by both? Cap the count |
The time-to-fill row deserves particular attention if you operate in India. Measuring to offer acceptance produces a flattering number that bears no relationship to when work actually gets done, because a 90-day notice period sits between the two. Report both, or report to start date and explain why it looks slow.
Where ISO 30414 fits — and where it does not
If you want an external anchor for your definitions, ISO 30414 is the one that exists. The second edition, published in August 2025, is titled Human resource management — Requirements and recommendations for human capital reporting and disclosure, and restructures the 2018 original into required and recommended disclosures with closer alignment to sustainability reporting frameworks including IFRS S1, ESRS and India's BRSR.
Sources disagree on the metric count. Some describe 58 metrics across 11 areas; others describe 69 across 11 areas, split into 14 required and 55 recommended. We have not obtained the standard itself, which is paywalled, and are reporting the disagreement rather than resolving it. If the exact figure matters to you, buy the standard.
The honest guidance for a mid-market company is narrow: do not adopt ISO 30414 as a board reporting framework. It is a disclosure standard designed for external reporting and regulatory convergence, and a board pack built from it will be comprehensive and unread. What it is genuinely good for is definitions — if you need a defensible answer to "how are you calculating that," borrowing a standardised definition is stronger than inventing one, and it makes your numbers comparable if you later face disclosure obligations.
Those obligations are moving in one direction. The SEC has required material human capital disclosure in annual filings since November 2020, IFRS S1 took effect in January 2024, and SEBI's BRSR regime already imposes workforce disclosure on India's largest listed companies. A private mid-market company is not in scope today. A company planning an IPO in three years is building the data trail now whether it intends to or not.
Four presentation rules
Trend, not point. One number is an anecdote. Eight quarters is a signal. Choose the single most important metric and give it the trend line; the rest can be value, prior, plan.
Variance against plan, always. A board approved a plan. Every metric that relates to it should be expressed against it.
State the assumption in the same sentence as the number. "At 50% of CTC, which is the conservative end of the published range, attrition cost us four crore" invites scrutiny of a defensible assumption. The same figure without the qualifier invites dismissal of the whole exercise.
Exceptions by name, or the word None. A compliance percentage tells a board nothing actionable. A named late filing with a named owner and a date does.
Why this is usually harder than it should be
Nine metrics sounds like a modest ask, and for most mid-market companies producing them reliably takes a week. The reason is rarely analytical. Headcount and plan sit in a spreadsheet, exits and reasons in the HRMS, requisition and offer dates in the recruiting system, cost to company in payroll, filing status in a compliance tracker. Every metric above is a join across at least two of those, and the joins are done by matching names in exported files.
That is why the definitional discipline in this article matters more than it should have to. When the same fact exists in four places, the definition is effectively decided by whichever export somebody used that quarter — and the number changes without anyone choosing to change it.
Helion holds hiring, HR and payroll on one database, so a departure carries its own date, reason, cost and requisition history without a reconciliation step. That does not tell you which nine metrics to report. It does mean the answer to "why is this number different from last quarter's" is never "because we pulled it differently."
Frequently asked questions
How many HR metrics should a board pack contain?
Fewer than ten, on one page. Nine is a workable ceiling covering staffing, retention, cost and compliance. The discipline is in what you exclude.
Should engagement survey results go to the board?
Only with an action attached. A score that moves without explanation or consequence teaches a board to skim your section.
What is the most commonly missing metric?
Regretted attrition. Total attrition without it cannot distinguish a retention problem from healthy turnover, and boards routinely draw the wrong conclusion from the aggregate.
Should we report against external benchmarks?
Sparingly, and only where you can name the source and its methodology. Your own trend against your own plan is more defensible and more useful than a comparison a director cannot verify.
Do we need ISO 30414?
Not as a reporting framework at mid-market scale. Borrow its definitions if you want external grounding, particularly if disclosure obligations are foreseeable for you.
The nine-metric selection, the page layout and the presentation rules are Helion's framework, offered as judgement rather than as survey findings or a standard. ISO 30414:2025 details are from secondary reporting; sources conflict on whether the second edition contains 58 or 69 metrics and we have not obtained the standard, which is paywalled. Regulatory references to SEC human capital disclosure, IFRS S1 and SEBI BRSR are contextual and not a statement of your obligations — confirm scope with your advisors. All illustrative figures are fictional. Current as at July 2026. General information, not advice.