You put budget into Meta every month and applications come in. The question your agency actually wants answered is a different one: what does a candidate really cost you, and what does a placement cost? Below you work through the whole chain, from ad budget to a signed placement, with a fully worked example and the levers that genuinely move that number.
What “cost per candidate” actually means
Cost per candidate is not one number but a family of numbers, and they easily differ from each other by a factor of ten. Before you can compare anything, you have to pin down which number you mean.
- Cost per application — ad budget divided by the number of submitted forms. The easiest to measure and the least meaningful.
- Cost per qualified candidate — only applicants who meet your basic requirements and can actually be reached.
- Cost per intake interview — candidates who genuinely showed up for a conversation.
- Cost per placement — budget divided by the number of candidates who start work. This is the number that touches your margin.
If someone tells you they bring in candidates for €12, ask what the €12 buys. An application from someone who never answers the phone does not cost you €12; it costs €12 plus half an hour of recruiter time.
The chain from budget to placement
Between your ad budget and a placement sit six multiplications. Every step has its own percentage, and every percentage shrinks what is left.
- Budget and CPM determine how many impressions you buy.
- CTR determines how many of those impressions become clicks.
- The conversion rate of your vacancy page determines how many clicks become applications.
- Your qualification criteria determine how many applications are serious candidates.
- Your follow-up speed determines how many of those candidates show up for an intake.
- The client side determines how many intakes turn into placements.
Because it is a multiplication, improvements compound: lift two steps by 20% each and the same euro produces 44% more placements. Anyone who works through their recruitment funnel step by step usually sees within half an hour where the money leaks away.
Worked example: a €2,000 budget from impression to placement
Assume the following. Note that these are assumptions to demonstrate the arithmetic, not industry figures.
- Monthly budget: €2,000
- CPM: €8
- CTR: 1.2%
- Vacancy page conversion rate: 8%
- 35% of applications meet your basic requirements
- 50% of those qualified candidates show up for an intake
- One in six intakes leads to a placement
€2,000 at a €8 CPM buys 250,000 impressions (2,000 divided by 8, times 1,000). Of those, 1.2% are clicked: 3,000 clicks, or €0.67 per click. Your vacancy page converts 8% of those clicks: 240 applications, €8.33 per application. 35% of them are qualified: 84 candidates, €23.81 per qualified candidate. Half attend an intake: 42 intakes, €47.62 per intake. One in six is placed: 7 placements, €285.71 per placement.
What that number is worth
€286 per placement means nothing on its own. It only becomes a decision next to your gross margin. If that margin runs to around €2,500 over the life of a placement, then €2,000 of media budget sits against roughly €17,500 in margin. If it is €400, this is a completely different conversation. And media budget is not your only cost: screening and calling 240 applications takes recruiter hours that belong in the same sum.
Why cost per candidate is the wrong yardstick on its own
Almost every intervention that lowers your cost per application can raise your cost per placement. That is not theory, it is arithmetic.
Say you shorten your form from nine fields to three. Conversion rises from 8% to 12%: the same 3,000 clicks now produce 360 applications at €5.56 each — a third cheaper. But because you ask for less, the share of qualified applicants drops from 35% to 25%: 90 qualified candidates instead of 84, so €22.22 against €23.81. Still progress, but far smaller than the headline suggests. Had that share fallen to 20%, you would be left with 72 candidates at €27.78: more expensive than where you started, with 120 extra files to call through.
“A cheaper application is only a gain if it keeps the same odds of becoming a placement.”
The levers that move the number
Creative
The ad moves two variables at once: your CTR and, through Meta’s auction weighing expected engagement, your effective CPM as well. In the worked example: take CTR from 1.2% to 1.8% and the same budget buys 4,500 clicks instead of 3,000, dropping cost per application from €8.33 to €5.56. Video shot at a real workplace with a real employee structurally outperforms a polished vacancy banner here.
Audience breadth
The narrower your targeting, the smaller the auction and the higher your CPM: you pay a premium to keep reaching the same people. Buying broadly and letting the ad do the filtering often works better. Name the job title, the region and the salary in the first three seconds and the wrong people stop clicking by themselves.
Form length and where you qualify
You qualify somewhere: in the ad, on the page, in the form or on the phone. Move it earlier and volume falls while quality rises; move it later and you pay in recruiter time. Where that balance sits depends on how many calling hours your team has free. The biggest gain here comes from a vacancy landing page that converts without pulling in the wrong people.
Response speed
The most underrated lever is not in Ads Manager. Candidates who apply through social tend to do it in the evening and usually at more than one agency; whoever calls first speaks to someone who has committed to nothing yet. If attendance falls from 50% to 30%, those same 84 qualified candidates leave you roughly 25 intakes and just over 4 placements. Your cost per placement jumps from €286 to about €480, with nothing changed in the campaign.
The application-to-intake ratio
This is the step where most agencies lose money and the only one that sits entirely inside your own walls. Measure it per recruiter and per week. If it varies sharply between people, your problem is not your budget but your follow-up process.
Why engineering, healthcare and logistics produce different numbers
Sector differences are real, but they follow from four mechanisms: the size of the audience on the platform, how many parties bid on those same people, how heavy the qualification requirements are, and how much margin a single placement produces.
- Engineering — a small, sharply defined audience that many agencies bid on at once. Higher CPMs and lower volumes, but usually a margin per placement that carries it.
- Healthcare — wide reach, but hard requirements such as diplomas and professional registration. Application volume can be high while your qualification rate stays low; the gain lies in filtering earlier.
- Logistics — a broad audience and a low barrier to entry, so plenty of volume at a low cost per application. Against that stand shorter placements and more drop-off before the first working day.
Never adopt another agency’s numbers as a norm; a CPM from a LinkedIn post is an anecdote as far as you are concerned. Run your own campaign for six to eight weeks, record your own chain, and use that as your baseline.
Without measurement you are calculating with guesses
UTMs
Set one fixed convention across every ad link: source, medium, campaign, and a distinct value in utm_content per creative. Without that last one you know that Meta works, but not which video does the work.
Conversion events
Fire your conversion on the thank-you page, not on the submit button click — otherwise you also count everyone who hit a validation error. Separate “application started” from “application completed” and send the latter server-side through the Conversions API where possible.
Source registration in your ATS
This is where the chain closes. Pass the source through as a hidden field and let it travel into your ATS; never ask the candidate for it in a dropdown, because that gets clicked at random. Each month, tie your ad spend to the placements that came from that source. This instrumentation is a standard part of how we set up JAM campaigns; what that involves is on our services page.
Conclusion
Cost per candidate is a perfectly good working number, as long as you know it is an interim figure. The decision your agency makes hangs on cost per placement measured against your margin. Work through the chain once from end to end with your own figures, record the six percentages, and then check each month which percentage moved. Usually it is not the CPM everybody stares at, but the step between application and intake.
Want to know what your chain produces and where the biggest leak sits? Book a call and we will work through your funnel together with your own numbers.
