How to measure recruitment performance and ROI with limited access to data and analytics

Most talent teams get asked for recruitment ROI about a week before the budget meeting, and that is usually the week they find out the numbers were never captured. No applicant-source field. No hire dates in one place. A cost per hire that lives in someone's head. The good news is that you can measure recruitment performance and ROI with a spreadsheet, five metrics and roughly 30 minutes a month, as long as you stay honest about what a thin dataset can and cannot prove.

That honesty matters more than the tooling. A number nobody can defend is worse than no number at all, because it gets quoted in a board deck and then used to cut the wrong budget line.

TL;DR

• Recruitment ROI is the business value a hire creates minus what it cost to find them, divided by that cost. The formula is simple; the inputs are what break.
• Only 13% of UK employers surveyed measure recruitment ROI at all, so a team with a spreadsheet and five metrics is already ahead of most of the market.
• Five metrics carry almost all the signal: cost per hire, time to hire, offer acceptance rate, 12-month new-hire retention, and hiring manager satisfaction.
• Where data is missing, use a stated proxy and label it as one. Write the assumption down next to the number so nobody inherits a fake precision.
• Thin data can prove direction and rank options. It cannot prove causation, and claiming it does is the fastest way to lose the room.

What is recruitment ROI?

Recruitment ROI is the value a hire delivers to the business minus the cost of hiring them, expressed as a percentage of that cost. Value usually means productivity, revenue contribution or cost avoided. Cost means everything spent to fill the role: advertising, agency fees, tooling, and the hours your own team put in.

The definition is easy to agree on and hard to operate, because "value" is a forecast and "cost" is scattered across three systems that do not talk to each other. Most teams stall right there and quietly go back to reporting time to fill.

Almost nobody finishes the job. Just 13% of employers surveyed for the CIPD resourcing and talent planning survey say their organization measures the ROI of its recruitment processes, and the split is uneven: 18% of large employers against 8% of small and medium ones, 16% in the private sector against 5% in the public sector and 6% in non-profits. The same survey found the three methods that do get used are cost per hire, new-hire turnover and new-hire performance, which is a useful signal in itself. The teams already doing this are not running anything exotic. They are running the basics consistently.

Read that number the other way round and it is an opportunity. If your hiring managers think measurement requires a people analytics function, the market evidence says otherwise, because most of the organizations that have one are not measuring this either.

How do you calculate recruitment ROI?

Recruitment ROI = (value of the hire minus total recruitment cost) divided by total recruitment cost, times 100. Total cost is external spend plus internal spend. Value is whatever your finance team already accepts as a per-head contribution figure, so borrow theirs rather than inventing one.

Here is a deliberately ordinary example. A 40-person accounting firm hires six seasonal associates. Job ads cost 1,200 dollars, the practice manager spends 22 hours screening at an internal rate of 45 dollars an hour, and one hire comes through a 3,000 dollar agency fee. Total cost is 5,190 dollars, or about 865 dollars per hire. Finance already books a seasonal associate at 14,000 dollars of billable contribution over the season. Six of those is 84,000 dollars. ROI works out near 1,518%.

That number is not a claim about the firm's hiring quality. It is a claim about one season, six people and one assumption borrowed from finance. Written that way, it survives a challenge. Written as "our recruiting delivers 15x", it does not.

The UK picture gives you a sanity check on the cost side: the same CIPD survey puts median recruitment cost at around 2,000 pounds for senior managers and 1,500 pounds for other employees. Those are UK medians across all sectors, so treat them as a range check rather than a target. If your figure lands far outside them, the input is probably wrong before the conclusion is.

Why does recruitment data go missing?

It is rarely negligence. Data goes missing because the system of record was chosen for something other than reporting, and nobody was ever given the job of keeping it clean.

• The source field is optional. Optional fields get skipped under load, and hiring is always under load.
• Costs sit in finance, not in recruiting. Agency invoices land in accounts payable with no link back to the requisition.
• Internal time is never logged. Interviewer hours are the single largest hidden cost in most hiring processes and almost nobody counts them.
• People leave and take context with them. The recruiter who knew why that role took 90 days is now at another company.
• Post-hire outcomes live somewhere else entirely. Performance and retention data sits with HR operations, on a different review cycle, in a different tool.

This is not a small-team problem either. Fewer than 10% of companies can correlate their HR and people data to business metrics at all, based on Josh Bersin's people analytics research from late 2024. Large enterprises with dedicated analytics functions are mostly in the same position as a 40-person firm with a spreadsheet. They just have prettier dashboards on top of the same gaps.

Which recruitment metrics matter most?

Five. Cost per hire, time to hire, offer acceptance rate, 12-month new-hire retention, and hiring manager satisfaction. Together they cover cost, speed, market competitiveness and quality, and each one can be captured by hand without any analytics tooling.

The temptation is to track 25 metrics because a list somewhere said you should. Resist it. A team that reports five numbers every month for a year has something a team that reported 25 numbers once does not: a trend line.

If you want the fuller list once the habit sticks, these recruitment metrics worth tracking extend the same logic, and the yield ratio is the next one most teams add.

How do you measure performance without analytics?

Use proxies, and label every one of them. A proxy is a stand-in you can defend: interview hours estimated from calendar invites instead of logged time, or source of hire captured by a single question on the application form. The rule is that the assumption travels with the number, in the same cell.

Three techniques do most of the work here.

1. Capture forward, not backward. Do not spend three weeks reconstructing last year. Start a clean sheet this month and accept that your first honest trend line is 90 days away.
2. Borrow finance's numbers. Internal hourly rates, per-head contribution and loaded salary costs already exist somewhere in the business. Using finance's figure makes the result defensible in the meeting where it matters.
3. Sample instead of counting everything. You do not need interview hours for all 40 hires. Time five of them properly, use the average, and say so in a footnote.

External benchmarks fill the rest of the gap. The Bureau of Labor Statistics JOLTS data for July 2026 put US job openings at 7.3 million against 5.1 million hires, with 3.1 million quits and a quits rate of 1.9%. That tells you whether a slow month was your process or the market, which is a question your own data can almost never answer alone.

Pro tip: keep one tab called Assumptions and nothing else. Every proxy, every borrowed rate, every sample size goes there with a date. When somebody challenges a number nine months later, that tab is the difference between a two-minute answer and a lost afternoon.

How do you measure quality of hire?

Combine one retention signal with one performance signal and one manager signal, then hold the definition still for at least four quarters. Quality of hire is not a single score you can buy. It is a definition your organization agrees to and then stops renegotiating, which is the part most teams skip.

A workable version with no analytics tooling: did the person still work here at 12 months, did they reach the agreed ramp milestone, and would the hiring manager hire them again. Three fields. Scored quarterly.

The cost of getting this wrong is not subtle. Replacing one employee runs between one-half and two times their annual salary, and Gallup's turnover cost analysis puts voluntary turnover at roughly 1 trillion dollars a year for US businesses. For a 100-person organization on an average salary of 50,000 dollars, that is 660,000 to 2.6 million dollars annually. Quality of hire is the only recruitment metric that moves that number.

This is also where pre-hire evidence starts to earn its place. The Testlify Quality-of-Hire Learning Model connects pre-hire evidence with post-hire outcomes such as ramp time, retention and manager feedback, then refines assessments and benchmarks based on what actually predicted success. It is a methodology rather than an automatic proof: it needs your post-hire outcome data to close the loop, and without that data it stays a good habit rather than a measurement.

What can thin data actually prove?

Direction and ranking. Not causation. A thin dataset can tell you that referrals convert better than job boards this quarter and that your offer acceptance rate has fallen for three months running. It cannot tell you that a new assessment step caused a retention improvement, because you have no control group and your sample is 30 people.

That distinction protects you. The teams that get burned are the ones that presented a 40% improvement from an 11-person sample and then had to walk it back. Say "the direction is consistent across three quarters" and you keep credibility. Say "we proved it" and you spend the next meeting defending arithmetic.

There is a second limit worth naming. Recruitment data ages badly because the market moves underneath it. The World Economic Forum Future of Jobs research found 63% of employers name the skills gap as the main barrier to business transformation, with close to 40% of job skills expected to change. A cost per hire benchmark from two years ago is describing a different labor market than the one you are hiring in.

So treat your own trend line as the primary evidence and external benchmarks as context. Not the reverse.

And there is a case for not doing this at all yet. If your team fills fewer than about 10 roles a year, every metric on the list has a sample size too small to read, and the effort is better spent writing down what good looks like for each role. Measurement earns its keep at volume, or when a specific decision depends on it, such as whether to renew an agency contract or move budget from job boards to referrals. Measuring because a template said to is how teams end up with a spreadsheet nobody opens.

Build the measurement habit in 90 days

The first month is setup, the second is capture, the third is when the numbers start meaning something.

1. Days 1 to 15. Agree the five metrics and write one sentence defining each. Get the hiring managers to agree to the definitions before any data is collected, not after.
2. Days 16 to 30. Build one spreadsheet with a row per requisition and the Assumptions tab. Pull internal hourly rates from finance.
3. Days 31 to 60. Capture live. Add the 30-day hiring manager question to your existing onboarding email so it costs nobody extra work.
4. Days 61 to 90. Run the first review. Report the direction of each metric, the sample size, and the two assumptions you are least confident about.
5. Day 90 onward. Change one thing per quarter and watch what moves. One change at a time is the only way thin data can attribute anything.

Teams that already run an applicant tracking system can skip most of the manual capture, though the cost and post-hire fields still usually need filling by hand. For the wider picture, measuring talent acquisition ROI covers the funnel above the requisition, and hiring metrics at enterprise scale covers what changes when the volume goes up.

Hire on evidence, not guesswork

Structured, scored assessments give you a pre-hire signal you can actually compare across candidates, which is the input the whole measurement exercise depends on. Testlify runs skills, cognitive and role-based assessments with multi-reviewer scoring, integrates with the applicant tracking system most teams already run, and ships a basic tracking layer for teams that have none. Start a free trial or book a walkthrough to see how the scoring data comes out.

FAQs