Human Resources KPI Dashboard Example

Human Resources Performance Dashboard

It's the budget meeting in October, and the controller has a fair question. Headcount is the same as it was a year ago. So why did HR make 102 hires?

The national numbers wouldn't have raised an alarm. The Bureau of Labor Statistics JOLTS release for August 2026 put the quits rate across all US nonfarm employers at 1.9%, with total separations at 3.2%. But the same survey's industry table shows manufacturing quits at 1.6% that month, and between 1.2% and 1.5% in every month from January 2025 to July 2026. A plant that measures itself against the national figure has more room to drift than it thinks.

People are staying put, too. BLS's employee tenure survey for January 2026 found median tenure of 4.1 years, up from 3.9 in 2024, and 4.9 years in manufacturing. The share of wage and salary workers with a year or less at their employer fell from 22.2% to 20.6%. When the labor market is that settled, a rise in quits among your newest people says more about your first few weeks than about the economy.

The example board tracks an invented employer: a US maker of commercial kitchen equipment with about 410 people. Its monthly figures are modeled, not a customer's. Three targets below have a published benchmark behind them, and the table says which. The rest are starting points.

What is a Human Resources KPI Dashboard?

Overview

An HR KPI dashboard tells you whether the company is hiring the people it needs, keeping them past the first few months and getting them to work, at a cost the business can carry.

It starts with the requisition. Time to fill says how long a role sat open. Offer acceptance rate says how many of the people you chose said yes. Cost per hire says what each outside hire cost to find, and internal fill rate says how many openings went to someone already on the payroll.

Then the first ninety days. 90-day new hire turnover counts the people who left before their third month was out. It's the number that ties recruiting and onboarding together, and one that many HR dashboards leave off. Training hours per employee shows whether new starters, and everyone else, are actually being trained.

Then the workforce as a whole. Voluntary turnover rate tracks the people who chose to leave against average employee headcount. Employee absence rate tracks the hours lost to sickness and unplanned time off. The employee engagement score, taken from a short monthly pulse survey, is a read on how people feel about the place, a few months before some of them decide to leave.

Who benefits from this dashboard?

1. HR Manager: Voluntary turnover and 90-day turnover on one screen show whether the people leaving are long-serving staff or last quarter's hires, and those call for very different responses.
2. Recruiter: Time to fill and offer acceptance show where the process stalls. 90-day turnover is the honest check on whether a fast fill was a good one.
3. Plant Manager and Supervisors: Absence and voluntary turnover explain the overtime bill. A supervisor whose second shift is carrying three open roles feels it before HR's report does, and internal fill rate shows whether their best operators are getting the lead and setup jobs.
4. Controller or CFO: Cost per hire, read with the hire count it is divided by, shows what turnover costs in cash, not just in disruption.
5. Training Lead: Training hours per employee show whether onboarding still runs at full length when the plant is busy, which is exactly when it tends to get cut.

Don't celebrate a falling cost per hire until you've checked how many hires it was divided by. Recruiting has large fixed costs: recruiters' pay, the applicant tracking system, job board subscriptions. Hire the same jobs twice and those costs get spread over twice as many people, so the cheapest hires on the board can be the ones you had to replace.

Dashboard Example Overview

The example company builds commercial kitchen equipment in the Southeast: one plant on two shifts, with about 300 hourly production and warehouse staff and about 110 in engineering, sales, field service and the office. HR is a team of four, two of them recruiters. It reports ten KPIs monthly:

• Average Employee Headcount: Headcount at the start of the month plus headcount at the end, divided by two. The denominator for the voluntary turnover and training figures, and the check against the headcount plan.
• Internal Fill Rate: Positions filled by people already on the payroll, through promotion or transfer, divided by all positions filled in the month, inside and outside. Hourly line seats nearly always go to outside hires, so the rate lives on lead, setup, maintenance and office roles. ISO 30414, the human capital reporting standard, lists it among its recruitment, mobility and turnover metrics as the percentage of positions filled internally.
• Time to Fill: Average calendar days from requisition approval to offer accepted, for roles filled in the month. Approval, not posting, because roles often wait a week for a manager to write the ad.
• Offer Acceptance Rate: Offers accepted divided by offers made. A drop usually means pay, shift or the candidate's view of the job, and each needs a different fix.
• Cost per Hire: Internal and external recruiting costs in the month, divided by employees hired. Internal covers recruiters' pay and the applicant tracking system. External covers job boards, agencies, background checks and referral bonuses. Employees hired, the denominator, counts new starts in the month: rehires count, internal transfers don't. It feeds this KPI rather than getting a tile of its own.
• 90-Day New Hire Turnover: Hires who left within 90 days of their start date, divided by all hires in the six monthly start groups whose 90 days ended in the last six months. Six groups instead of one, because nine hires a month is too few to read on its own.
• Voluntary Turnover Rate: Voluntary quits in the month divided by average headcount. Layoffs, dismissals and retirements are tracked separately. That matches the way JOLTS counts quits, so the JOLTS quits rate is the fair comparison.
• Employee Absence Rate: Unplanned hours absent divided by scheduled hours. Vacation and approved leave are left out. This is the hours-based measure, the closest match to BLS's lost worktime rate.
• Employee Engagement Score: Favorable answers as a share of all answers on a five-question monthly pulse survey, run by shift.
• Training Hours per Employee: All recorded training hours in the month, onboarding included, divided by average headcount.

The two years go like this.

In October 2024 the board is quiet. Voluntary turnover is 1.5% against a 2.0% target, set a year earlier from the national JOLTS quits rate. 90-day new hire turnover is 12.5%. Internal fill rate is 27.3%, mostly operators stepping up to lead and setup jobs. Training hours per employee are 2.4, most of it two-week onboarding for new starters.

In December the plant's onboarding trainer retires and isn't replaced. Onboarding shrinks to three days of safety training and paperwork, and second-shift hires go straight to the line. Training hours drop to 0.9 that month, and everyone reads it as the holidays. From January to October 2025 they sit between 1.2 and 1.5, and the monthly review notes that hiring has been light.

Voluntary turnover never does anything dramatic. It reads 1.7% in March 2025, then 2.0% in April and May, exactly on target, so the gauge stays green. June hits 2.4%, July and August 2.2%, and summer is when quits always rise. September is back to 2.0%. Meanwhile recruiting looks better than ever. Time to fill has fallen from 38 days to 30, and cost per hire is $3,605 in August and September, the lowest of the two years.

Internal fill rate slides on the same timetable, from 22.2% in January 2025 to 10.0% in April and 8.3% in August and September, one internal move a month. With new starters reaching the line half trained, supervisors stop releasing their experienced operators for lead and setup openings, and those jobs go to outside candidates. Nobody decides that. Each supervisor just says not this month.

The controller's question sends the HR manager to a line graph instead of the gauge. Twelve months of voluntary turnover on one line, with a trend line through it, shows what no single month did. The six months to March 2025 averaged 1.4%. The six months to September averaged 2.1%. Against manufacturing's 1.2 to 1.5%, the plant had been high since March. HR then pulls start and leaving dates from the HRIS and adds 90-day new hire turnover to the board, backfilled. It has risen in nine of the eleven months since, from 12.5% to 24.0% in September 2025. One in four new hires is gone inside three months.

The fix, from November 2025, is dull. A second-shift supervisor takes on onboarding half-time, and the two weeks come back, with a named buddy on the line. Supervisors hold a short check-in at 30, 60 and 90 days. Second-shift candidates work a paid four-hour shadow shift before an offer is made. Training hours go to 2.3 in November, and from January 2026 they stay between 2.4 and 3.1. With trained replacements arriving again, supervisors let operators move up: internal fill rate is 15.4% in November 2025 and 25.0% by April 2026.

The improvement is slow. 90-day turnover keeps rising after the fix, to 29.5% in January 2026, because it's still counting people hired before November. Then it eases: 27.1% in March, 23.6% in May, 18.9% in July and 15.8% in September 2026. That's still above the 12.5% where it started. Voluntary turnover for April to September 2026 averaged 1.75%, against 2.1% for the same months of 2025. The target is now 1.6%, taken from the manufacturing figure, and the plant isn't there yet.

Some of what the company gave up was deliberate. Time to fill went back up, to between 37 and 39 days from March 2026, because the shadow shift adds a step. Offer acceptance fell from 91.7% in September 2025 to 80.0% in February 2026, as candidates saw the second shift up close and some said no. The company's bet is that many of them would have quit in week three. Cost per hire is $4,332 in September 2026, higher than a year earlier and still below SHRM's nonexecutive average. Absence didn't follow the plan either: 2.9% in January 2026, the worst month on the board, was flu season, and nothing on this dashboard would have prevented it. The employee engagement score, which slid from 71% to 65% while second shift covered for empty roles, was back to 70% by August 2026. And internal fill rate reached 30.8% in September 2026, but every operator who moved up left a line seat to backfill, so promoting from within didn't bring hiring down.

List of Human Resources KPIs

Cost per hire, voluntary turnover and absence rate have published US figures to compare against, named in the table. The other targets are this example's starting points, so replace them with your own once you have a year of history.

Why Human Resources Dashboards Matter

HR numbers are slow and small. At a plant of 410 people, a bad month is two extra resignations, and a monthly report can explain away two resignations every time. The damage comes from the months that each looked fine.

Several of these KPIs also pull against each other. Time to fill and cost per hire both improve when screening gets lighter, and 90-day turnover pays for it a quarter later. Offer acceptance goes up when candidates see less of the job, and falls when you show them the night shift. Training hours cost production time, which is why they get cut first in a busy quarter. Internal fill rate depends on them: a supervisor only lets a good operator move up when the replacement arrives trained. Each trade can be the right call. It's only a problem when nobody sees it being made.

For a manager, that means reading the recruiting numbers with a 90-day lag in mind, and reading turnover against your own industry instead of the national rate. Gallup's 2025 survey found 31% of US employees engaged at work, the same as 2024 and down from 36% in 2020, so most employers can't count on goodwill to carry a new hire through a rough first month.

For a manufacturer, the practical version is simple. Every quarter, list the people who left before day 90, with their shift, their supervisor and their start date. If one shift or one month of starts keeps appearing, you've found your onboarding problem, months before voluntary turnover turns red. For the production side of the same plant, see the manufacturing dashboard example, and for the agencies that fill roles like these, the staffing agency example.