KPI Examples for HR: 25 Metrics That Matter

What are the most important KPI examples for HR?

The most important HR KPIs are the ones tied to a decision someone will actually make: whether to change a hiring process, keep a manager, fund a training program, or approve a headcount plan. A practical starter set is time to fill, cost per hire, voluntary turnover, regretted attrition, eNPS, and time to productivity.

Everything else on this list is context. That is not a criticism — context is what stops you from optimizing the wrong thing. But a dashboard with 25 live metrics produces meetings about definitions instead of decisions.

How should you choose from this list?

Choose by audience, not by completeness. The executive team needs six numbers that connect people to business outcomes. HR operations needs a different six that show process health. A hiring manager needs three.

A workable selection rule:

• Five to seven metrics per audience. More than that and nobody owns any of them.
• At least one leading indicator. Time to fill is lagging by the time you measure it; offer acceptance predicts it, and pipeline health predicts offer acceptance.
• One cost metric. HR that cannot state a cost per outcome will keep losing budget arguments to teams that can.
• One quality metric. Speed and cost without quality produces fast, cheap, wrong hires.

The table below is the full set, grouped by the question each family answers.

How do the hiring KPIs work in practice?

Hiring metrics are the most measured and most misunderstood group, because each one is trivially gameable in isolation.

Time to fill measures the requisition’s life: approval to accepted offer. SHRM’s benchmarking work puts the median at about 44 days for non-executive roles, which is a useful sanity check rather than a target. Some organizations measure to the start date instead; pick one convention and keep it. A specialist engineering role and a front-desk role should not share a target.

Time to hire starts later — at first candidate contact — and is shorter. The gap between the two is your process overhead: approvals, job posting, scheduling. That gap is usually where the fixable delay lives.

Cost per hire follows the SHRM/ANSI convention: all internal and external recruiting costs divided by hires in the period. SHRM’s widely cited average is close to $4,700, with executive and technical roles far above it. The number is only comparable if your cost base is consistent, so write down what you include before you publish it.

Offer acceptance rate is the cheapest early warning you have. Below roughly 80% usually means compensation, process speed, or employer brand — in that order. It is also the metric most often improved by making interviews slower and more thorough, which then damages time to fill. Watch the pair.

Source effectiveness answers “where do good hires come from?” not “where do most applicants come from.” A job board may deliver 60% of applications and 15% of hires; a referral program may deliver 8% of applications and 35% of hires. Only one of those deserves more budget.

> Common mistake: Reporting time to fill as an average across all roles. A single executive search can hide eight fast hires. Report the median and the 80th percentile, and segment by role family.

How do you calculate the cost of a bad hire or a departure?

You calculate turnover cost by adding the costs that actually change when someone leaves: recruiting the replacement, the vacancy gap, onboarding, and the productivity lost while the new person ramps.

• Recruiting: agency fee or internal sourcing time, advertising, interview time
• Vacancy: lost output or overtime for the team covering the work
• Onboarding: training hours, materials, manager time
• Ramp: the gap between a new hire’s output and a fully productive employee, usually 2–4 months

Gallup’s frequently cited estimate is that replacing an employee costs between one-half and two times their annual salary, which is a useful range for a sanity check — not a substitute for your own calculation.

Example 1: cost per hire across a mixed hiring year

A 240-person company made 11 hires last year. Recruiting spend:

• Agency fees: 3 hires at 18% of a $70,000 salary = $37,800
• Job boards and advertising: $9,400
• Referral bonuses: 4 × $1,500 = $6,000
• Recruiter time allocated: 0.4 FTE at $78,000 = $31,200, of which 60% went to these roles = $18,720

Total: $71,920 ÷ 11 = $6,538 per hire, well above the ~$4,700 benchmark — and the agency line explains most of the gap. Cutting two agency hires in favour of referrals at the same 18% rate would save $25,200 minus $3,000 in bonuses, or $22,200, dropping cost per hire to about $4,520.

Example 2: what a vacancy actually costs

Three revenue-facing roles sit open for 44 days each. Each role contributes about $2,400 of gross margin per week. At 6.3 weeks, the vacancy cost is 3 × 6.3 × $2,400 = $45,360.

Cutting time to fill to 30 days (4.3 weeks) saves 2 weeks per role: 3 × 2 × $2,400 = $14,400. That is the number to put next to a request for an extra recruiter or a scheduling tool — not “we will hire faster.”

Example 3: the cost of a two-point turnover improvement

Headcount is 240, annual voluntary turnover is 17%, and the average salary is $64,000. Exits: 240 × 17% = 40.8 per year. At a conservative replacement cost of 0.75× salary ($48,000), the annual bill is 40.8 × $48,000 = $1,958,400.

Reducing turnover to 15% cuts 4.8 exits, saving 4.8 × $48,000 = $230,400 a year. Note the sensitivity: if your true multiplier is 1.5× rather than 0.75×, the same reduction is worth $460,800. State your multiplier and defend it.

Which retention metrics actually predict loss?

Retention metrics predict loss only when they are split. A single turnover percentage mixes early quits, planned retirements, and regretted losses of people you wanted to keep.

Voluntary turnover is the base rate. Calculate it monthly and annualize only for comparison, never for trend — a bad quarter annualized four times is how HR loses credibility with finance.

Regretted attrition is the number that matters to the business: exits you would have prevented. It requires a manager rating at exit, which is subjective, so define “regretted” before the conversation, not after.

Early attrition — exits within twelve months — is the cleanest signal of a hiring or onboarding problem. If it is above 15–20%, the issue is usually role realism or onboarding, not compensation.

High-performer retention is the metric that should reach the board. Losing five average performers is a cost; losing two people who carry critical knowledge is a capability risk.

Turnover cost turns all of the above into money. It is the metric that gets budget approved, provided the assumptions are written down.

> Practical tip: Report turnover with a cohort view once a quarter — hires by start month, still employed at 3, 6, 12 months. It turns a single flat percentage into a picture of where people leave.

What do engagement KPIs tell you that turnover does not?

Engagement metrics are leading indicators: they move before people resign. Turnover tells you what happened; engagement tells you what is forming.

eNPS comes from one question — how likely are you to recommend this company as a place to work, on a 0–10 scale. Promoters score 9–10, detractors 0–6, and the score is promoters minus detractors as a percentage. Common reading: above 0 is acceptable, 10–30 is healthy, 50+ is exceptional.

Engagement index aggregates several survey items into a % favorable figure. It is more stable than eNPS and less quotable, which is often a good trade.

Manager effectiveness is the highest-leverage slice of any engagement survey. People leave managers more often than they leave companies, so track the manager items separately and report them by team — with a minimum group size to protect anonymity.

Internal mobility and absence rate complete the picture. Low mobility suggests a blocked career path; rising absence often precedes turnover by a quarter.

Example 4: calculating eNPS from a real response set

120 employees respond. 46 rate 9–10, 33 rate 0–6, and 41 are passives.

• Promoters: 46 ÷ 120 = 38.3%
• Detractors: 33 ÷ 120 = 27.5%
• eNPS: 38.3 − 27.5 = 10.8, reported as 11

That sits at the bottom of the healthy band. The more useful output is the split by department: if one team of 18 reports −22 while the rest average +18, you have a local problem, not a company culture problem.

How do you set targets without inventing them?

Set targets in three layers: your baseline, your peer or market reference, and your structural constraint.

Start with the baseline — the last four to eight periods, not one. Then apply an external reference where a credible one exists: SHRM’s 44-day median time to fill, the ~$4,700 average cost per hire, the 10–30 eNPS band. Finally, check the structural constraint: a 12-person HR team cannot deliver a 30-day time to fill for licensed engineers, whatever the benchmark says.

A target should be a range with an agreed response. “Time to fill 38–44 days; above 50 triggers a process review with the hiring manager” is a working KPI. “Time to fill: 30 days” is a wish.

Three escalation rules keep targets honest:

1. The metric leaves the agreed band for two consecutive periods.
2. Any single team or role family is more than 50% worse than the company figure.
3. A metric improves while its paired quality metric declines — the classic gaming signature.

What mistakes make HR KPIs useless?

• Tracking everything. Twenty-five live metrics means twenty-five things nobody owns.
• Averaging away the story. Means hide outliers; use medians and percentiles.
• No denominator discipline. Cost per hire with a shifting cost base is not comparable across quarters.
• Measuring activity instead of outcome. Training hours are an input; capability is the output.
• Ignoring the pair. Speed without quality, engagement without turnover, cost without retention.
• Publishing without a response plan. A metric with no trigger and no owner is decoration.

Where does Doitify fit?

Most HR KPIs depend on project work being visible: onboarding plans, hiring sprints, training rollouts, and retention initiatives all run as projects with owners and deadlines, and the data usually lives in three different places.

Doitify holds tasks, sub-tasks, checklists, owners, due dates, and work reports in one workspace, which helps in two concrete ways. Onboarding becomes a checklist per new hire with a completion rate you can report at day 30 and day 90. Hiring and retention initiatives become projects with milestones, so “we are working on time to fill” turns into a tracked plan with a deadline and an owner.

Transparency: Doitify is our product. It is a project and team management platform, not an HR information system, and it will not replace your payroll or applicant tracking data. It helps when the question is “is the work actually happening, and who owns it?”

Conclusion

The 25 KPI examples for HR above are a menu, and the discipline is in choosing badly on purpose: six numbers per audience, each with a formula, an owner, a cadence, and a written response when it moves. Time to fill and cost per hire explain your hiring engine. Regretted attrition and early attrition explain whether it is producing people who stay. eNPS and manager effectiveness tell you what is forming before it shows up in a resignation letter.

Pick one metric from each family this quarter, calculate it from last year’s data, and publish it with its assumptions. The second quarter is where the value appears — not because the number is new, but because the conversation around it finally has a shared language.

If this post on KPI examples for HR was helpful, you might also enjoy Gantt Chart Examples for Project Management and Project Management Software for Hybrid Teams: 2026 Guide.