Which Recruitment Advertising Platform Gives the Best Return on Ad Spend? A Worked Example

No recruitment advertising platform gives the best return on ad spend for every employer. The platform with the best return is the one that produces the most hires, or qualified applicants, per dollar for your roles, measured in your applicant tracking system rather than in the vendor's dashboard. That is a number you can calculate yourself, and the worked example below shows why the platform with the cheapest clicks often delivers the worst return once you count what happens after the click.

What does return on ad spend mean in recruitment?

Return on ad spend, or ROAS, comes from ecommerce advertising. Google Ads describes target ROAS as the average conversion value you want for each dollar spent, and gives the example that $5 in sales for every $1 of ad spend is a 500% ROAS.

Recruitment has no sale at the end of the funnel, so the conversion value has to be assigned. Google makes the same point for any advertiser whose conversions are not purchases: conversion values let you "learn the total business value that Google Ads generated, rather than simply the number of conversions that have happened."

For hiring, the practical formula is:

Recruitment ROAS = (hires × value of a hire) ÷ ad spend

Most teams find the value of a hire hard to agree on, so they track the inverse instead: cost per qualified applicant and cost per hire. Both answer the same question. A platform with a lower cost per hire at the same quality has a higher return.

How do you put a value on a hire?

Pick one definition, write it down, and use it for every platform you compare. Three are common.

1. Avoided cost. What you would otherwise pay to fill the role, for example a staffing agency fee or overtime to cover the empty shift.

2. Benchmark cost. An external reference point. SHRM benchmarking data put the average cost per hire at nearly $4,700, which covers all recruiting costs, not only advertising.

3. Productive value. The contribution of a filled position per week, multiplied by how many weeks sooner the hire starts.

The definition matters less than keeping it constant. If one vendor's case study values a hire at the agency fee and another values it at annual salary, their ROAS figures cannot be compared.

Which platform wins in a worked example?

Assume an employer spends $20,000 on the same set of warehouse associate openings with three platforms, one at a time, under the same conditions. Every input below is an assumption for illustration, not a benchmark. The hire rate from a qualified applicant is held at 1 in 6 for all three, because the same recruiters run the same interviews.

• Platform A bills per click at $0.80 and buys the cheapest traffic it can find.

• Platform B bills per click at $1.60 and targets a narrower audience.

• Platform C bills per qualified applicant at $45.

The $3,000 value per hire is an assumed avoided cost, the fee this employer would otherwise pay a staffing partner. Change it and every ROAS figure moves in proportion, but the ranking stays the same.

Platform A shows the best number on its own dashboard, a cost per click half that of Platform B. It delivers the worst return, because its cheap traffic applies less often and qualifies less often. Platforms B and C end up almost level on cost per hire. The difference between them is who carries the risk if the funnel underperforms, which the next sections cover.

Which inputs change the answer most?

Run the same sensitivity test on Platform A to see where return actually comes from.

A ten point improvement in the share of applicants who qualify is worth more than a 20% discount on media. This is the main reason ROAS comparisons built on cost per click mislead: they measure the price of the input, not the yield of the funnel. Audience quality, the length of the application flow, and the screening questions decide most of the result.

Why can you not compare platform dashboards directly?

Each platform reports on the unit it bills and counts it with its own rules. Three differences distort side by side comparisons.

Different counting systems. Clicks, leads and applications are often counted by the channel or the vendor, not by you. Ask whose count is authoritative before you compare any two reports.

Different attribution rules. Google Ads explains that "attribution models let you choose how much credit each ad interaction gets for your conversions," and that a last click model "gives all credit for the conversion to the last-clicked ad." A candidate who sees a social ad, later searches your company name and applies can be claimed by two platforms at once.

Different definitions of quality. One vendor's "qualified" can mean a completed form, another's a candidate who passed knockout questions. If the definitions differ, the cost per qualified applicant figures are not the same metric.

The fix is to measure every platform at the same point in the funnel, in the same system. For most employers that system is the ATS, where applications, screening outcomes and hires already live.

How do you run a fair ROAS test between platforms?

A short, controlled test answers the question better than any vendor benchmark.

1. Fix the value of a hire and the definition of a qualified applicant before the test starts, and put both in writing.

2. Use the same jobs and the same period for each platform, or split comparable locations between platforms if running them in parallel.

3. Tag every application at source so the ATS shows which platform produced each candidate.

4. Measure at the qualified applicant stage first. Hires are rare compared with applications, so a difference of a few hires between platforms can be chance. Qualified applicants give you more data points in the same time.

5. Read hires last, once enough have happened to compare, and calculate cost per hire and ROAS from ATS data, not from the vendor's report.

6. Check the billing unit against the outcome. If a platform bills per click, you carry the cost of every visitor who does not apply. If it bills per qualified applicant, the vendor carries it, but only if the qualification criteria are yours.

Does the pricing model change the return?

At the same funnel performance, no. Platform B and Platform C in the example deliver nearly the same cost per hire. The pricing model changes what happens when performance slips.

Suppose the application rate on Platform B falls from 9% to 6% in a tight labor market. Clicks still cost $1.60, applications drop to 750, qualified applicants to 300, and hires to 50. Cost per hire rises from $267 to $400, and ROAS falls from 1,125% to 750%. Under per qualified applicant billing, the same drop means fewer qualified applicants arrive for the budget, but each one still costs $45, so cost per hire stays at $270 if interview conversion holds. The employer's risk moves from price to speed.

The reverse risk exists too. If the screening questions are loose, more candidates qualify who are not worth interviewing, the hire rate falls, and cost per hire rises even though cost per qualified applicant looks stable.

Where does Wonderkind fit when you compare return on ad spend?

Wonderkind is a social programmatic job advertising platform that lets an employer choose the billing unit that matches the stage it wants to measure. Its pricing page sets out a performance budget in which the customer pays per click for "people who open your job," per lead for "applicants who start," or per qualified applicant for "candidates who pass screening." The platform is organized in three modules, Attract, Qualify and Deliver, with candidates delivered into the customer's ATS so they can be measured there.

Wonderkind's terms define a qualified applicant as a candidate who completes the application flow, passes all knockout questions and meets the job criteria "as configured by the Customer," and is delivered into the customer's ATS, with disputes raised within 10 business days. That puts cost per qualified applicant on a definition the employer controls. The trade-off is the one in the example above: the return depends on how strict those criteria are, and a loose set of questions produces billable applicants who do not get hired.

Frequently asked questions

What is a good ROAS for recruitment advertising?
There is no standard benchmark, because every employer values a hire differently. Compare platforms against each other using the same value per hire, and track whether your own cost per hire falls over time.

Is cost per click a good way to compare recruitment advertising platforms?
No. Cost per click measures the price of traffic, not the number of qualified applicants or hires it produces. In the worked example, the platform with the lowest cost per click had the highest cost per hire.

Should I measure ROAS on applications or on hires?
Measure on qualified applicants during the test, because they accumulate fast enough to compare, then confirm on hires once enough have happened.

Can I trust a vendor's reported ROAS?
Treat it as a starting point. Recalculate it from your ATS data, with your own value per hire and your own definition of a qualified applicant.

Does paying per qualified applicant guarantee a better return?
No. It moves the cost of drop off to the vendor, but the return still depends on the criteria you set and on how many qualified applicants your team converts into hires.