Most articles about recruitment ROI publish a formula that is not a return on investment. Some divide an employee’s lifetime value by the cost of hiring and retaining them, which produces a benefit-to-cost ratio. Others divide total new-hire salary by recruitment spend, which produces figures in the thousands of percent. Both will get you removed from a CFO’s meeting.

This guide gives you the correct formula, two worked examples with the arithmetic shown, a defensible way to build the benefit term, and the international standards that actually define these metrics. It also names two widely quoted figures you should stop using, including one attributed to the US Department of Labor that appears in no Department of Labor publication anyone has ever located.

A disclosure first. The previous version of this page published the lifetime-value-divided-by-cost formula. It was wrong, and the correction is below. We would rather fix it in public than quietly swap it out.

Key Takeaways

• ROI is (net benefit minus investment) divided by investment, expressed as a percentage. A ratio of value to cost is a benefit-to-cost ratio, not an ROI, and presenting one as the other overstates your result by 100 percentage points.
• The hard part is the benefit term, not the arithmetic. Four benefit components are genuinely measurable: cost of vacancy avoided, cost per hire reduced, early attrition avoided, and a performance differential. A hire’s total output is not one of them.
• There are international standards for these metrics and almost nobody cites them. ISO/TS 30407 covers cost per hire, ISO/TS 30411 covers quality of hire and publishes an actual formula, and ISO/TS 30410 covers impact of hire including the impact of unfilled critical positions.
• Time to fill and time to hire are different metrics. Only time to fill drives cost of vacancy, and conflating them is the most common measurement error in this area after the formula itself.
• Cost of vacancy is highly sensitive to one assumption. In the worked example below, changing how much work colleagues absorb flips the ROI from plus 269 percent to minus 100 percent. Any calculator that does not let you set that assumption is not telling you the truth.
• Honest recruitment ROI claims are modest. Peer-reviewed research puts the validity of even the strongest selection method at around .42, which accounts for roughly 18 percent of the variance in later job performance. Onboarding, management and role design account for most of the rest.
• Stop quoting “a bad hire costs 30 percent of first-year earnings, according to the US Department of Labor.” No such publication has been identified, and no source repeating it provides a document reference.

The Formula Most Guides Publish Is Not an ROI

Wrong Formula One: Lifetime Value Divided by Cost

Stated as: an employee’s lifetime value, divided by the cost to hire plus the cost to retain them. It appears on competing pages for this keyword and it appeared here. Six things are wrong with it:

• It is a benefit-to-cost ratio, not a rate of return. A benefit-to-cost ratio of 1.0 is breakeven. An ROI of 0 percent is breakeven. So presenting a ratio as an ROI overstates the result by 100 percentage points, every time.
• It books gross value rather than net. ROI measures the return after recovering the investment. This formula never subtracts it.
• The time horizons do not match. A multi-year lifetime value divided by a one-off cost, with no discounting, compares two different things.
• There is no counterfactual. The candidate you would otherwise have hired would also have produced value. Only the difference is attributable to your recruitment process.
• “Lifetime value” has no accounting definition for an employee, and for most roles revenue is not attributable per head at all.
• The cost to retain does not belong in a recruitment denominator. Salary is payment for work delivered, not an investment in recruiting.

Wrong Formula Two: New-Hire Salary Divided by Recruitment Cost

Stated as: number of hires multiplied by average salary, divided by recruitment spend. This treats the salary you pay as a return you receive, which inverts the sign of a cost. It is also why you occasionally see recruitment ROI quoted at 20,000 percent, a figure no finance function will accept.

The Correct Recruitment ROI Formula

The standard definition, applied to recruitment:

Recruitment ROI = (Benefit attributable to the recruitment investment, minus the investment) divided by the investment, multiplied by 100

Two rules make it defensible:

• The numerator is the incremental benefit, meaning the value created because of the recruitment investment, not the total value of the people hired. If you would have filled the role anyway, the benefit is the difference your investment made, not the whole outcome.
• The denominator is the recruitment investment only. Sourcing, recruiter cost, agency or RPO fees, technology, assessments and hiring-manager time. Not salary, not benefits, not onboarding, unless you are explicitly measuring the ROI of onboarding instead.

Stated that way, recruitment ROI stops being a number you can make arbitrarily large and becomes a claim you have to defend. That is the point of it. It also has to sit inside a wider plan, which is what a talent acquisition strategy is for.

Building a Defensible Benefit Term

This is where the real work is, and where almost every published guide stops. Four benefit components can actually be measured and defended.

1. Cost of Vacancy Avoided

If your investment reduced time to fill, the benefit is the days saved multiplied by the daily cost of that vacancy, multiplied by the number of hires. This is usually the strongest component because the mechanism is direct: the role was empty for fewer days. Methodology and its weaknesses are set out below.

2. Cost Per Hire Reduced

The cleanest component of all, because both sides are cash. If your cost per hire fell from a measured baseline, the benefit is the reduction multiplied by hires. The only discipline required is that the baseline must be measured on the same cost definition as the new figure, which is where the standards below earn their keep.

3. Early Attrition Avoided

If better selection or a better candidate experience reduced the share of hires leaving within twelve months, the benefit is the reduction in departures multiplied by the cost of replacing each one. Defensible, provided you use your own replacement cost rather than a figure from an article.

4. A Performance Differential

The honest version of “we hire better people”. The benefit is the performance difference between the hires you made and the hires you would otherwise have made, over a defined period. There is an established body of work in industrial and organizational psychology on estimating this, generally called utility analysis, which models the value of a selection method as a function of its validity, the number of hires, their expected tenure and the variability of performance in the role.

Two honest observations about it. First, it requires an estimate of how much performance varies in currency terms, which most organizations do not have. Second, it is the only approach that models the counterfactual properly, which is why it is also the only credible way to claim a quality benefit. If you are going to claim one, this is the framework to use, with help from someone who has done it before.

What is not a benefit term: the total output, revenue contribution or salary of the people you hired. Those would have existed under almost any process that filled the role.

The Investment Side: What Counts as Recruitment Cost

You cannot compute a ratio without agreeing the denominator, and this is where internal teams and agencies talk past each other. There are published standards for exactly this, and no competing page on this keyword cites any of them.

• ISO/TS 30407:2018, Human resource management, Cost-per-Hire. The international technical specification for the metric, published in 2018 and current.
• ANSI/SHRM 06001.2012, Cost-per-Hire, an American National Standard developed with SHRM and approved in February 2012. Worth noting that we have not been able to confirm whether it has since been revised or withdrawn, so check its current status before citing it in a document that matters.

The approach both describe is the same, and it is simple:

Cost per hire = (Total internal recruiting costs + Total external recruiting costs) divided by the number of hires in the same period

Internal costs	External costs
Recruiter and HR compensation, loaded for employer taxes and benefits	Agency and search fees
Hiring manager and interviewer time	Job advertising and job board spend
Recruiting technology, including ATS and sourcing tools	Assessments and background checks
Employee referral bonuses	Relocation and candidate travel
Allocated overhead where your policy includes it	Sign-on bonuses

The line that matters most is hiring manager and interviewer time. Almost no internal team loads it in, which means almost every internal cost per hire is understated, which in turn means almost every internal-versus-agency comparison is wrong in the same direction. A standard that requires you to include it is doing you a favor, not creating paperwork.

On benchmarks: SHRM has published a figure of nearly 4,700 US dollars as an average cost per hire. Treat it carefully, because SHRM’s own article giving that figure states no year and no sample size. It is useful as an order of magnitude and not as a target.

Worked Example One: Agency or In-House

The most common real recruitment ROI question. A company hires 40 people a year at an average salary of 80,000, and is deciding whether to keep using contingency agencies or build an internal team. All figures are illustrative but the arithmetic is complete.

The Agency Baseline

Contingency fees are typically published in the range of 15 to 30 percent of first-year base salary, commonly 20 to 25 percent for professional roles. Note that this range comes from recruitment firms publishing their own pricing rather than from an independent survey, so treat it as a typical published range and not a benchmark.

At 20 percent: 80,000 times 0.20 equals 16,000 per hire. Across 40 hires that is 640,000 a year. For how the model works more generally, see what is a recruitment agency.

The In-House Investment, Costed to the Standard

Cost line	Calculation	Annual
Two recruiters	70,000 base each, loaded at 1.3 times for employer taxes and benefits	182,000
ATS and sourcing tools	Licences and seats	30,000
Job advertising	Boards and paid posts	25,000
Assessments and background checks	150 per hire, 40 hires	6,000
Referral bonuses	2,000 each on 8 referred hires	16,000
Hiring manager and interviewer time	12 hours per hire, 40 hires, at 75 per loaded hour	36,000
Total recruitment investment		295,000
Internal cost per hire	295,000 divided by 40	7,375

The ROI

• Investment: 295,000
• Benefit: 640,000 of agency fees avoided
• Net benefit: 640,000 minus 295,000 equals 345,000
• ROI: 345,000 divided by 295,000 equals 117 percent

And now the caveat that makes the number honest. That calculation assumes the internal team delivers the same 40 hires, at the same quality, in the same time. That assumption is where in-house transitions usually fail. If the internal team delivers 32 hires rather than 40, you still pay agency fees on the remaining 8, which adds 128,000 and cuts the ROI to roughly 73 percent. If time to fill also lengthens, the cost of vacancy on every role erodes it further.

So the useful output of this exercise is not “in-house wins”. It is that in-house wins if it delivers the same volume at the same speed, and the business case should be written with that condition stated and monitored. See RPO vs in-house recruitment for the fuller comparison.

Worked Example Two: Reducing Time to Fill

The same company invests 60,000 in a sourcing tool, a process redesign and a part-year coordinator, and reduces average time to fill from 50 days to 38 days. Twelve days saved per hire, across 40 hires.

Step One: Build the Daily Cost of Vacancy

Assume the company has revenue of 120 million and 300 employees, giving revenue per employee of 400,000. Over a 260-day working year that is 1,538 a day of gross revenue attribution per head.

Two adjustments that most calculators omit:

• Absorption. Colleagues absorb some of the work. Assume half the output genuinely does not get done: 1,538 times 0.50 equals 769 a day of real lost output.
• Salary not paid. While the role is empty you are not paying for it: 80,000 divided by 260 equals 308 a day saved, which must be netted off.

Net daily cost of vacancy: 769 minus 308 equals 461 a day.

Step Two: The ROI

• Benefit per hire: 12 days times 461 equals 5,532
• Benefit across 40 hires: 221,280
• Investment: 60,000
• Net benefit: 221,280 minus 60,000 equals 161,280
• ROI: 161,280 divided by 60,000 equals 269 percent

Step Three: The Sensitivity Test Nobody Publishes

Now change one assumption. Suppose colleagues absorb 80 percent of the work rather than 50 percent, which is entirely plausible in a well-staffed team:

• Lost output becomes 1,538 times 0.20 equals 308 a day
• Net of the 308 salary saving, the daily cost of vacancy is zero
• The benefit is zero, so the ROI is minus 100 percent

One assumption, which nobody measures, moves the answer from plus 269 percent to minus 100 percent. That is the single most important thing to understand about recruitment ROI, and it is why a calculator that does not expose the absorption assumption is not giving you an answer, it is giving you its author’s opinion.

What to do about it in practice: state your absorption assumption explicitly, use a conservative figure, and present the ROI as a range rather than a point estimate. A CFO will trust a range with stated assumptions far more than a precise number they cannot interrogate.

Cost of Vacancy: The Method and Its Weaknesses

The live version of this page did not cover cost of vacancy at all, and almost no competing page covers it properly either. Here is the method and an honest account of what is wrong with it.

The Method

Cost of vacancy = (daily lost output + daily coverage cost + daily team productivity loss, minus daily salary not paid) multiplied by days vacant

• For revenue-generating roles, use the role’s own quota or territory revenue divided by working days, rather than a company average. A salesperson’s vacancy cost is their quota, not the firm’s revenue per head.
• For non-revenue roles, substitute team productivity loss: the number of affected colleagues, multiplied by the share of their productivity lost, multiplied by their fully loaded daily cost. Add coverage costs such as overtime, contractors and manager time.
• Working days are conventionally taken as 260 a year, sometimes 250.
• Revenue per employee, where you use it as a proxy, is a defined benchmarking measure maintained by APQC. Use your own revenue divided by your own headcount rather than an industry median.

There is also a standards hook here that is unique on this topic: ISO/TS 30410:2018 on the impact of hire includes a clause on the impact of unfilled critical positions. It is the only international standard that addresses vacancy cost at all, and it is worth knowing it exists when you are defending the method internally.

Five Legitimate Objections to It

If your CFO raises any of these, they are right to:

• Revenue per employee attributes joint output to individuals. Revenue is produced by people together with capital, brand, product and process. Dividing it per head implies a causal share that does not exist.
• Output is rarely lost one for one. Colleagues absorb work, deadlines slip rather than vanish, and some tasks turn out not to have needed doing. The absorption factor is the whole ballgame, as the example above shows.
• Unpaid salary is a real saving and most calculators omit it. Leaving it out systematically overstates vacancy cost, and it is the most common error in vendor calculators.
• For overhead roles the productivity loss percentage is an assumption, not a measurement. Say so when you present it.
• Cost of vacancy is a rate, so it scales with time. A long vacancy in a low-impact role can be made to look catastrophic simply by multiplying a small daily figure by a large number of days. Sanity-check the total against what the business actually noticed.

One more thing worth saying plainly: there is no published benchmark for cost of vacancy per day, and there cannot be one. Every per-day figure in circulation is a vendor’s illustrative example. It is entirely role-specific and company-specific, which means it has to be calculated rather than quoted.

Key Metrics to Measure Recruitment ROI

These are the inputs. Each one has a definition worth getting right, because a metric computed two different ways cannot be compared across periods or against anyone else.

Cost per Hire

Total internal plus external recruiting costs, divided by hires in the same period. Covered in full above, with the cost categories.

The discipline that matters: pick a definition, write it down, and keep it stable. A cost per hire that falls because you quietly stopped counting interviewer time has not improved.

Time to Fill, and Why It Is Not Time to Hire

These get used interchangeably and they measure different things:

• Time to fill runs from the requisition being approved or opened to the offer being accepted. It includes approval delays and sourcing lag.
• Time to hire runs from a candidate entering the pipeline to that candidate accepting. It measures only the candidate-facing part of the process.

Time to fill is always the longer of the two, and only time to fill drives cost of vacancy, because the role is empty during the approval and sourcing phases too. If your team reports time to hire and your finance function assumes it is time to fill, your vacancy cost is understated.

Measuring both is also diagnostic. A short time to hire inside a long time to fill tells you the bottleneck is upstream, in approvals or sourcing, not in your interview process. That is a different problem with a different fix. See the recruitment funnel for where the stages sit.

On benchmarks, a word of caution. Figures around 42 to 44 days are widely attributed to SHRM, but we could not trace them to a dated SHRM report, only to secondary restatements. Measure your own baseline and improve against it rather than against a number you cannot source.

Quality of Hire

The hardest metric in recruitment and the one that matters most, because a fast cheap hire who fails is not a win.

There is an international standard for it, which almost nobody uses: ISO/TS 30411:2018, Human resource management, Quality of hire metric. It sets out two approaches. The first measures performance against objectives, defined as the sum of performance ratings of new hires on completion of a review period, divided by the number of new hires, with the review period set at 3, 6, 9 or 12 months. The second measures satisfaction with the hire’s knowledge, skills and abilities against expectations. Notably, the specification explicitly excludes impact of hire and retention of hire, which are treated as separate metrics elsewhere in the standards family.

The practical approaches, each with its known failure mode:

Approach	What it measures	Its weakness
First-year performance rating	Manager assessment at 6 or 12 months	Rater leniency and central tendency. Ratings are rarely comparable across managers
Ramp time to productivity	Days until the hire reaches a defined output level	Requires a productivity baseline per role, which most organizations do not have
Retention at 12 months	Whether the hire is still there	Binary and lagging, and it conflates recruiting quality with management quality
Hiring manager satisfaction	Perception of the hire and the process	A perception measure, prone to halo effects and post-hoc rationalization
Composite quality index	A weighted blend of the above	The weights are arbitrary, so it is only comparable against itself over time

One fact worth knowing before you benchmark: no cross-industry quality of hire benchmark exists, because there is no shared definition to benchmark against. Any provider offering to compare your quality of hire to an industry average is comparing it to their own construct.

Offer Acceptance Rate

Offers accepted divided by offers extended. A falling acceptance rate usually points at compensation, candidate experience or the time your process takes.

One citable benchmark: the National Association of Colleges and Employers reported an overall offer acceptance rate of 69.3 percent in its 2022 Recruiting Benchmarks Report. Note the population carefully, because it is US college and university recruiting rather than general hiring, so it is a reference point rather than your target.

Diagnostically, offer acceptance is the cheapest metric to improve on this list. Pay benchmarking and a faster process both move it, and both are within your control. See global salary benchmarking and how to improve candidate experience.

Cost of Early Attrition

The share of hires leaving within their first year, multiplied by what it costs to replace them.

We are not quoting an early attrition benchmark, because we could not find one with a traceable methodology. Use your own rate. And use your own replacement cost, built from your own cost per hire plus your own cost of vacancy, rather than a figure from an article.

Source Effectiveness

Not on the live version of this page and worth adding, because it is the metric that tells you where to spend. For each channel, track hires produced, cost per hire, time to fill and twelve-month retention. The useful insight is almost always the same: the cheapest source per application is rarely the cheapest source per retained hire, and referrals usually win on retention while producing too few candidates to rely on alone. See passive candidate recruitment.

The Attribution Problem: Why Honest ROI Claims Are Modest

This section is the reason to trust the rest of this guide, and no competing page on this keyword includes it.

Attributing a new hire’s performance to your recruitment process is statistically weak, and there is peer-reviewed evidence for how weak. Sackett, Zhang, Berry and Lievens, writing in the Journal of Applied Psychology in 2022, revised long-cited estimates of selection method validity sharply downward from the figures that had been standard since 1998. On their revised estimates, structured interviews correlate with later job performance at around .42, work samples around .33 and cognitive ability tests around .31. We would recommend checking those coefficients against the paper itself before quoting them in a board document, since we are working from a summary rather than the original tables.

The arithmetic that follows matters. At a correlation of around .42, the selection method accounts for roughly 18 percent of the variance in subsequent job performance. The remaining majority is explained by onboarding, management quality, role design, team context and everything else that happens after the hire starts.

Three consequences you should build into any ROI claim:

• Do not book a hire’s full performance as a return on recruitment spend. That claims credit for variance recruiting did not produce, and anyone numerate will spot it.
• Lead with the benefit components where the mechanism is direct. Cost per hire reduction and cost of vacancy avoided are defensible because the causal chain is short. Quality claims require the counterfactual and should be presented more cautiously.
• Frame the benefit as a difference, not a total. The credible question is what the hire you made delivers relative to the hire you would otherwise have made, over a defined period.

This is not an argument against measuring recruitment ROI. It is an argument for presenting a smaller number you can defend rather than a larger one you cannot, which is also the difference between a budget that survives scrutiny and one that does not.

Two Figures You Should Stop Quoting

“A Bad Hire Costs Up to 30 Percent of First-Year Earnings, According to the US Department of Labor”

This appears on hundreds of HR pages, including competing pages for this keyword. We went looking for the underlying Department of Labor publication and could not find one. Searches restricted to the Department’s own domains returned nothing on point, and not one of the many secondary sources repeating it provides a document title, report number or year.

We are not claiming the underlying idea is wrong. Bad hires are expensive. We are saying that the specific figure has no identifiable source and the attribution appears to be invented, so do not put it in a business case. If a CFO asks where it comes from, “everyone says it” is not an answer.

“A Bad Hire Costs an Average of 14,900 US Dollars”

This one is real and traceable, which makes it usable, but only with its caveats stated.

It comes from a CareerBuilder release dated 7 December 2017, based on a survey conducted by Harris Poll among 2,257 hiring and human resource managers and 3,697 workers in the United States between 16 August and 15 September 2017.

The caveats, which almost nobody passes on:

• It is a self-reported estimate, not a measured cost. Respondents selected from bands rather than calculating their actual costs.
• The bands were mostly far lower than the average suggests. In the small-business cut, 17 percent chose 1,000 to 2,500 dollars and 20 percent chose 2,500 to 5,000, so the headline average is pulled up by a minority of large estimates.
• It is nearly a decade old and CareerBuilder does not appear to publish a comparable figure now.

What a Bad Hire Actually Costs, With a Real Methodology

If you want a sourced figure with a disclosed method, the best available is from the Recruitment and Employment Confederation’s 2017 report “Perfect match”, which modelled the cost of a poor mid-management hire on a 42,000 pound salary at 132,015 pounds, broken into 48,855 of direct costs, 19,355 of replacement cost and 83,160 of indirect costs.

Its method is disclosed: a YouGov survey of 501 UK HR decision-makers in April 2017, a ComRes survey of 206 employers, thirteen telephone interviews and a literature review. Two caveats to state if you use it: it is an illustrative scenario rather than a measured average, and the large indirect component is modelled rather than observed. The same report puts replacement costs generally at 50 to 150 percent of annual salary, which is a more useful planning range than a single figure.

Ways to Improve Recruitment ROI

Ordered by how reliably they move the number, rather than by how interesting they are.

1. Attack Time to Fill Upstream

If your time to hire is short but your time to fill is long, the problem is requisition approval and sourcing lag, not interviewing. That is usually the cheapest thing on this list to fix and it goes straight into the cost of vacancy term. Measure both metrics separately and you will see immediately which one to work on.

2. Automate the Manual Work

Scheduling, screening, status updates and reference collection consume recruiter hours that do not require judgment.

The ROI case here is straightforward because the benefit is recruiter capacity, which you can measure as hires delivered per recruiter. Be careful about the claim you make: automation reliably removes administrative time, and it does not reliably improve quality of hire. See recruitment automation and ATS in recruitment.

3. Reuse the Candidates You Already Have

Previous applicants, silver-medal candidates and past contractors are a sourced pool you have already paid for. Re-engaging them has close to zero marginal sourcing cost, which makes it the highest-ROI channel available to most teams.

The practical requirement is a database you can actually search, with consent and retention handled properly, which is a data protection question as much as a technology one. It is also one of the clearest differences between transactional recruiting and talent acquisition as a function.

4. Improve the Candidate Experience

A poor process loses candidates late, which is the most expensive place to lose them.

The measurable benefit is offer acceptance rate, which is why this is worth doing even though “experience” sounds soft. Faster decisions, clearer communication and fewer interview rounds all move it. See how to improve candidate experience.

5. Use Structured Interviews

Not on the live version of this page and it belongs near the top, because it is the best-evidenced intervention in the whole of selection. On the revised validity estimates cited above, structured interviews are the strongest commonly available predictor of job performance. Same interviewers, same questions, same scoring rubric, scored independently before discussion.

It costs almost nothing to implement, and it has the useful side effect of making your quality of hire data comparable across managers, which is a precondition for measuring it at all. See candidate screening.

6. Invest in Retention

The cheapest hire is the one you do not have to make. Reducing early attrition improves recruitment ROI directly, because every avoided departure is an avoided replacement cost.

Worth being precise about ownership. Early attrition driven by a mismatch between the role as sold and the role as experienced is a recruitment problem. Early attrition driven by management or compensation is not, and recruiting should not be held to account for it. Separating the two is what makes the metric useful. See employee compensation strategy and common HR problems.

7. Widen the Pool, Including Geographically

Hiring remotely or across borders enlarges the candidate pool, which usually shortens time to fill and often lowers cost.

The caveat is that it adds employment complexity, since you cannot employ someone in a country where you have no entity without an employer of record. Build that cost into the denominator rather than discovering it afterwards. See how to hire international employees and hiring remote developers.

8. Outsource Where the Unit Economics Favor It

Agencies, RPO and embedded recruiters all have a place. The question is which one is cheaper per retained hire at your volume, which the next section addresses.

Agency, In-House or RPO: The Unit Economics

Model	Typical pricing	Cost behavior	Where it wins
Contingency agency	Typically published at 15 to 30 percent of first-year base, commonly 20 to 25 percent, paid on placement	Purely variable. You pay nothing if nobody is placed	Low or unpredictable volume, hard-to-find roles, and when you need no fixed cost
Retained search	Commonly 25 to 33 percent or more, billed in thirds: engagement, shortlist, placement	Committed regardless of outcome. The retainer is usually non-refundable	Senior and confidential searches where you need guaranteed effort
In-house team	Fixed salaries plus tools, advertising and interviewer time	Almost entirely fixed. Cost per hire falls as volume rises and spikes when it drops	Steady, predictable volume above roughly 30 to 40 hires a year
RPO	Most commonly a fixed monthly management fee plus a cost per hire	Hybrid, which is the point	Variable volume, multi-country hiring, or when you need capacity without headcount

Two things worth knowing about these numbers. The agency and retained percentages are ranges that recruitment firms publish about themselves rather than independently surveyed benchmarks, so treat them as typical asking prices. On RPO, Staffing Industry Analysts reported in 2021 that roughly half of RPO contracts use a combination of fixed monthly fee plus cost per hire, with the cost-per-hire element often split equally between requisition opening and placement, and an FTE-plus-cost-per-hire model accounting for only about 2 percent of the market. Per-hire dollar ranges circulating for RPO are vendor marketing rather than published data.

The comparison most companies get wrong is agency fee against internal recruiter salary. Those are not comparable. An agency fee is a fully loaded cost of a delivered hire. An internal recruiter salary is one line in a cost base that also includes tools, advertising, assessments, referral bonuses and, above all, hiring manager and interviewer time. Cost them both to the standard definition above and the comparison becomes real. See is RPO right for you and global recruitment process outsourcing.

How TalentHero Helps You Improve Recruitment ROI

TalentHero recruits, employs and pays across Asia Pacific through our own entities. For recruitment ROI specifically, that matters for a reason that is easy to miss: when sourcing, employment and payroll sit with three different providers, the handoffs between them add days to time to fill and costs to the denominator that nobody owns.

• Recruitment on Demand. Embedded recruiting capacity priced against delivery rather than headcount, which keeps the cost variable while the volume is uncertain. See recruitment on demand.
• Employer of Record. Where you have no local entity, we employ your hires through ours, which removes the entity setup from the critical path and therefore from your time to fill.
• Global Payroll. Payroll run against current statutory rates and caps, so the loaded cost you modelled is the loaded cost you pay.
• Contractor Management. For genuinely independent engagements, which are sometimes the right answer and often cheaper.
• Workforce planning support. Including where widening the pool to a new market is the cheapest way to shorten a search. See hiring offshore tech talent in Southeast Asia.

And the thing most relevant to this page: we will cost a comparison honestly, including the hiring manager time on your side of the ledger, rather than presenting a fee against a salary and calling it a saving.

When a Different Answer Is Better

• High, steady volume in one country where you already have an entity. An internal team will beat any outsourced model on unit cost. Build it.
• A handful of genuinely rare senior searches a year. Retained search exists for a reason, and a contingency or embedded model is the wrong instrument.
• A footprint centered outside Asia Pacific. Our depth is regional, and we will say so early.
• You have not measured a baseline yet. Do that first. Any provider promising an ROI improvement against a baseline you cannot produce is promising something unmeasurable.

Takeaway

Recruitment ROI is worth measuring and easy to measure dishonestly. The discipline is in three things: use the real formula, build the benefit term from components where the causal chain is short, and state your assumptions so a finance colleague can argue with them.

If you want a starting point, measure three numbers this quarter: cost per hire to the standard definition including interviewer time, time to fill and time to hire as separate metrics, and twelve-month retention by source. Those three support most of the ROI claims worth making.

Get in touch with TalentHero and we will work through the numbers for your situation, including the cases where the honest answer is that your current setup is already the right one.

Frequently Asked Questions

1. What is recruitment ROI?

Recruitment ROI measures the return generated by money spent on recruiting, expressed as a percentage. The formula is the benefit attributable to the recruitment investment, minus the investment, divided by the investment, times 100. The critical word is attributable: the benefit is the value created because of the investment, not the total value of the people hired, because they would have produced value under almost any process that filled the role.

2. What is the recruitment ROI formula?

(Benefit minus investment) divided by investment, times 100. Avoid two common wrong versions. Dividing an employee’s lifetime value by hiring and retention costs gives a benefit-to-cost ratio, not an ROI, and overstates the result by 100 percentage points because a ratio of 1.0 is breakeven while an ROI of 0 percent is breakeven. Dividing total new-hire salary by recruitment spend treats a cost you pay as a return you receive.

3. What counts as the benefit in a recruitment ROI calculation?

Four components are defensible: cost of vacancy avoided through faster hiring, cost per hire reduced against a measured baseline, early attrition avoided multiplied by your own replacement cost, and a performance differential estimated properly through utility analysis. What does not count is the total output or salary of the people you hired, since that would have existed anyway.

4. What counts as recruitment cost?

Internal costs are recruiter and HR compensation loaded for taxes and benefits, hiring manager and interviewer time, recruiting technology, referral bonuses and allocated overhead. External costs are agency and search fees, job advertising, assessments and background checks, relocation and candidate travel, and sign-on bonuses. ISO/TS 30407:2018 is the international technical specification for cost per hire, and ANSI/SHRM 06001.2012 is the American standard. Interviewer time is the line most teams omit, which is why most internal cost per hire figures are understated.

5. What is the difference between time to fill and time to hire?

Time to fill runs from requisition approval to offer acceptance and includes approval and sourcing lag. Time to hire runs from a candidate entering the pipeline to that candidate accepting. And time to fill is always longer, and only time to fill drives cost of vacancy because the role is empty during the earlier stages too. Tracking both is diagnostic: a short time to hire inside a long time to fill tells you the bottleneck is upstream.

6. How do you calculate the cost of a vacancy?

Daily lost output plus coverage costs plus team productivity loss, minus the salary you are not paying, multiplied by days vacant. For revenue-generating roles use the role’s own quota rather than company revenue per head. For other roles use affected colleagues times productivity lost times their loaded daily cost. Two things most calculators get wrong: they omit the salary saving, and they assume all output is lost when colleagues absorb much of it.

7. Is there a benchmark for cost of vacancy per day?

No, and there cannot be one. Every per-day figure in circulation is a vendor’s illustrative example. Cost of vacancy depends entirely on the role’s output, the company’s revenue per head, how much colleagues absorb and the salary not being paid, so it has to be calculated for your situation rather than quoted.

8. What is a good cost per hire?

There is no universal answer, because it depends on role seniority, market and how honestly you cost it. SHRM has published an average of nearly 4,700 US dollars, though SHRM’s own article giving that figure states no year or sample size, so treat it as an order of magnitude. A more useful approach is to measure your own figure on a stable definition, segment it by role family and seniority, and track the trend.

9. How do you measure quality of hire?

ISO/TS 30411:2018 gives two approaches: the sum of new hires’ performance ratings at the end of a review period of 3, 6, 9 or 12 months divided by the number of hires, or satisfaction with their knowledge, skills and abilities against expectations. In practice teams use first-year performance ratings, ramp time to productivity, twelve-month retention, hiring manager satisfaction or a composite. Each has a known weakness, and no cross-industry benchmark exists because there is no shared definition.

10. Can you really attribute a hire’s performance to recruitment?

Only partly, and this is the honest limitation of recruitment ROI. Peer-reviewed research published in 2022 revised selection validity estimates downward, putting structured interviews, the strongest commonly used method, at around .42. That accounts for roughly 18 percent of the variance in later job performance, which means onboarding, management, role design and team context explain most of the rest. Book the full performance of a hire as a recruiting return and a numerate reader will discount everything else you said.

11. Does a bad hire really cost 30 percent of first-year salary?

That figure is almost always attributed to the US Department of Labor and we could not find any Department of Labor publication containing it. None of the many sources repeating it provides a document title, number or year. Bad hires are genuinely expensive, but use a sourced figure instead. The Recruitment and Employment Confederation’s 2017 “Perfect match” report modelled a poor mid-management hire on a 42,000 pound salary at 132,015 pounds, with its survey methodology disclosed, and put replacement costs generally at 50 to 150 percent of annual salary.

12. Is it cheaper to recruit in-house or use an agency?

In-house is usually cheaper per hire above roughly 30 to 40 hires a year, provided the internal team delivers the same volume at the same speed. The worked example above shows a 117 percent ROI on building in-house against 20 percent agency fees, falling to about 73 percent if the team delivers 32 of the 40 hires rather than all of them. The comparison only works if you cost interviewer and hiring manager time into the internal figure, which most teams do not.

13. How does RPO pricing work?

Staffing Industry Analysts reported in 2021 that roughly half of RPO contracts use a fixed monthly management fee plus a cost per hire, with the per-hire element often split equally between requisition opening and placement. Other structures include fixed cost per hire varying by source, project fees for volume hiring, and an FTE-based model that accounts for only about 2 percent of the market. Per-hire dollar figures quoted publicly are vendor marketing rather than published benchmarks.

14. What should I measure first if I am starting from nothing?

Three metrics. Cost per hire on a written definition that includes interviewer time. Time to fill and time to hire as separate numbers. Twelve-month retention broken down by source of hire. Those three support most defensible ROI claims, and crucially they give you the baseline without which no improvement can be demonstrated at all.