What is benchmarking?

Benchmarking is comparing your performance or your practices against a reference point: your own history, a competitor, an industry standard, or an organization that does one process unusually well. The point is to find the gap and then to find out what causes it.

Types of benchmarking

Benchmarking is classified on two axes, and most confusion comes from mixing them.

By comparison partner:

By object of comparison:

• Performance benchmarking compares outcome numbers. It tells you that a gap exists.
• Process benchmarking compares how the work is done: steps, cycle times, handoffs. It tells you why.
• Strategic benchmarking compares operating models and long-run capability choices.

Performance benchmarking is by far the most common in HR and the least useful on its own. A number without the process behind it produces a target nobody knows how to hit.

The practice was formalised at Xerox from 1979, where the copier business compared its unit manufacturing costs against Japanese competitors, then extended the method to non-competitors. Robert C. Camp’s Benchmarking: The Search for Industry Best Practices That Lead to Superior Performance (ASQC Quality Press, 1989) is where the four types and the ten-step process were codified.

The HR metrics people benchmark, and their formulas

Two definitional notes that cause most benchmarking arguments. Time to fill and time to hire are different measures and time to hire is always the shorter one. Retention rate is not the complement of turnover, because it excludes people hired and lost inside the same period.

Where a published standard exists, use it. Cost per hire has an ANSI-approved American National Standard sponsored by SHRM and approved in June 2012, which specifies exactly which costs belong in the numerator.

For external human capital reporting, ISO 30414:2025 replaced the 2018 edition in August 2025 and adds auditable requirements plus a mandatory baseline set of metrics. Most HR content still cites the withdrawn 2018 version.

External reference points worth using

• Quits and separations. The US Bureau of Labor Statistics JOLTS release for July 2026 put the quits rate at 1.9% and the total separations rate at 3.2%. This is the standard external comparator for a turnover rate.
• Absence. BLS Current Population Survey annual averages put the 2025 absence rate for full-time wage and salary workers at 3.2%, of which 2.2% was illness or injury.
• Pay movement. The BLS Employment Cost Index for the twelve months to June 2026 showed total compensation up 3.4%, wages and salaries up 3.2%.
• Engagement. Gallup’s State of the Global Workplace 2026 put global engagement at 20% and manager engagement at 22%.
• Recruiting difficulty. SHRM’s 2026 Talent Trends Report, published April 2026 from 2,094 HR professionals, found nearly 70% struggling to recruit for full-time roles and 42% reporting difficulty retaining them.

How to run a benchmarking exercise

1. Define the metric before you source the data. Write the formula, the numerator, the denominator and the exclusions. Most benchmarking failures are definitional, not analytical.
2. Set your internal baseline first. You cannot interpret an external gap without knowing your own trend and your own variance between teams.
3. Choose comparators deliberately. Same industry is not the same as comparable. Headcount, geography, seniority mix and business model all move these numbers more than management quality does.
4. Normalise. Adjust for size, sector and region, and check that the comparator’s denominator matches yours.
5. Find the gap and then find the cause. This is the step that separates benchmarking from data collection. Performance data tells you the gap size; only process comparison tells you what to change.
6. Set a target you can explain. A target copied from a benchmark without an accompanying process change is a wish.
7. Re-measure on a fixed cadence. Quarterly for operational metrics, annually for structural ones.

Common benchmarking mistakes

• Comparing definitions, not performance. Two organizations counting cost per hire differently will differ by more than any real gap.
• Chasing the median. A benchmark is a description of what others do, not a statement of what is good.
• Ignoring internal variance. If your best and worst teams differ by 30 points, the external benchmark is the less interesting number.
• Benchmarking a process you should abolish. Making a redundant step faster is still waste.
• Single-year comparisons. Most HR metrics are noisy at annual resolution and only mean something as a trend.
• Copying the target without the practice. The number is the output. The practice is the thing worth stealing.

How Engagedly helps

Benchmarking starts with clean internal data, which is where most HR teams stall. Engagedly’s Performance Suite holds goals, check-ins, feedback and review completion in one system, so internal comparisons across teams, managers and locations come from the same definitions rather than from four spreadsheets.

That internal baseline is what makes external comparison meaningful, since you can see your own variance before you go looking for a peer number.

Explore the Performance Suite or book a demo.

Benchmarking FAQs