How Much Should You Budget for Job Ads on Facebook and Instagram?

Most teams starting job ads on Facebook and Instagram pick a budget by feel. A few hundred per role, per month, split across every location with a vacancy. That number is usually wrong in both directions at once: too small per ad set for Meta's delivery system to work properly, and too large in total for the number of hires it produces.

The budget is not a marketing decision. It is arithmetic that runs backwards from the hires you need, with one hard floor set by how Meta's auction learns.

What is the right budget for job ads on Facebook and Instagram?

The right budget is the amount that produces enough of your chosen conversion event each week for the delivery system to optimize on, multiplied by the number of separate audiences you genuinely need. Everything else follows from those two numbers.

So the first question is not "how much can we spend" but "what are we counting". A small budget still produces plenty of clicks per week. The same budget might produce three or four qualified applicants who reach your ATS, which is neither enough signal for the system to improve nor enough volume for you to tell whether the campaign works.

How do you work the budget back from the number of hires you need?

Start at the end and divide. Four conversion rates sit between spend and a signed contract, each specific to your roles, markets and application flow.

1. Hires you need in the period. Take the requisition count, not the headcount plan.

2. Offer acceptance rate. If you need 20 starters and 4 in 5 offers are accepted, you need 25 offers.

3. Interview to offer rate and qualified applicant to interview rate. These two together tell you how many qualified applicants you must deliver. At a 40 percent qualified applicant to interview rate and a 50 percent interview to offer rate, 25 offers needs roughly 125 qualified applicants.

4. Cost per qualified applicant. This is the only number in the chain you buy directly.

The unit matters more than the rate, which is why it should be defined in writing rather than assumed. Wonderkind's contract definitions set a qualified applicant as a candidate who completes the application flow, passes all knock out questions as configured by the customer, meets the job description criteria as configured by the customer, and is delivered into the customer's ATS. All four conditions, not three. A vendor whose definition stops at "completes the application flow" is selling you a different thing at a similar price.

Work the same chain with clicks if that is what you buy, remembering that a click counts when the channel's own attribution system records it, so the billed number is the channel's count, not yours. The full cost per hire calculation sits one step further out again.

Why does the learning phase set the real minimum?

Meta's delivery system treats a new or recently edited ad set as unstable until it has seen enough results to predict outcomes. The Marketing API exposes this as a learning stage with three states: LEARNING, meaning "the ad set is still learning", SUCCESS, meaning "the ad set exited the learning phase", and FAIL, meaning "the ad set isn't generating enough results to exit the learning phase".

Two things follow directly from that third state.

First, a budget can be too small in a way that no amount of creative fixes. If an ad set produces a handful of conversion events per week, it sits in a state Meta itself describes as not generating enough results, and cost per result stays volatile.

Second, splitting a fixed budget across more ad sets makes this worse. Ten locations with a tenth of the budget each is ten ad sets that may never stabilize. Three regional ad sets are three that probably will.

The same API field counts conversions "since the time of its last significant edit during the learning phase". Editing targeting, creative or budget mid flight restarts the count. If you are going to change something, change it once, deliberately, and then leave it alone long enough to read the result.

Which conversion event should you pay to optimize for?

Pick the event closest to a hire that still occurs often enough to clear the learning problem above. In practice that is a trade off, and the three common units behave differently:

• Cost per click. Cheap, plentiful, and only loosely connected to hiring. Useful for learning which creative and audience combinations get attention.

• Cost per lead. A lead form submitted on the channel itself, for example a Meta, TikTok or Google lead generation form. High volume, low friction, and the qualification work still sits with your team.

• Cost per qualified applicant. The unit that matches what recruiters actually need, and the scarcest per euro of spend.

Whatever you choose, the reported number depends on the attribution window you set, which is a settings choice rather than a fact about candidate behavior. Google Ads makes the range explicit: a click through conversion window can be set anywhere from 1 to 30, 60, or 90 days with a 30 day default, engaged view defaults to 3 days, and view through defaults to 1 day. Two teams reporting different costs per application for the same campaign are often just using different windows.

Which bid strategy should you use for a job ad?

Meta documents four bid strategies, and the API names make the differences clearer than the interface labels do.

• LOWEST_COST_WITHOUT_CAP: "Facebook automatically bids on your behalf and gets you the lowest cost results." Start here. It finds the market rate for your audience without you guessing it.

• COST_CAP: "Get the most results possible while we strive to meet the cost per action you set." Move here once you know the cost per qualified applicant your business can carry. Meta states plainly that adherence to cost cap limits is not guaranteed.

• LOWEST_COST_WITH_BID_CAP: bids up as needed but never past your limit. Meta's own documentation notes this "requires more active bid management", which for most hiring teams means it is the wrong default.

• LOWEST_COST_WITH_MIN_ROAS: a value optimization option requiring a minimum return on ad spend floor. There is no revenue value on a hire, so this one does not apply to recruitment.

The practical sequence is to run lowest cost first, read the actual cost per result over a full week, then set a cost cap slightly above it rather than at the number you wish you were paying. A cap set below the market rate does not lower your costs. It stops delivery.

Should the budget sit at the campaign or the ad set level?

Meta's Advantage campaign budget is described as "a way of optimizing the distribution of a campaign budget across your campaign's ad sets", where the system "automatically and continuously finds the best available opportunities for results across your ad sets and distributes your campaign budget in real time".

For multi location hiring that is usually what you want, because it moves money toward the towns where people are actually applying instead of holding an equal amount in the ones where they are not. Three documented conditions decide whether you can use it:

• All optimization goals must be the same across ad sets.

• All ad sets share the same bid strategy, defined at campaign level, though bid amounts can differ per ad set.

• Above 70 ad sets with Advantage budgets enabled, you cannot edit the bid strategy or turn the feature off.

Keep budgets at ad set level when you have to guarantee spend to a specific client, region or role, which is the normal situation for staffing firms billing per client. Everywhere else, let the campaign move the money.

How does the employment declaration change the budget math?

Job ads on Meta must be declared in the employment special ad category, and the declaration removes most of the targeting that would normally make a small budget efficient. Meta's documentation states that for employment ads, age options are "generally fixed to include ages 18 through 65+", specific gender cannot be targeted, zips targeting is unavailable, location options are limited to city, metro area, small geo area, subneighborhood and electoral district with minimum radius requirements, excluding "any behaviors, demographics or interests is unavailable", and Lookalike Audiences are not available.

The budget consequence is direct. Your audience is blunter than a consumer advertiser's, so more of every impression reaches someone who will never apply, and the filtering targeting used to do has to happen in the creative and the application flow instead. Judge the campaign on cost per qualified applicant, because cost per click will look fine while the funnel leaks.

What does a worked example look like?

Twenty warehouse hires needed in a month, across three regions.

• 20 hires at an 80 percent offer acceptance rate means 25 offers.

• 25 offers at a 50 percent interview to offer rate means 50 interviews.

• 50 interviews at a 40 percent qualified applicant to interview rate means 125 qualified applicants.

• 125 qualified applicants across roughly 4.3 weeks is about 29 per week, or about 10 per week per regional ad set.

Ten conversion events per ad set per week is a workable floor, which is why three regional ad sets beat ten local ones at this volume. Multiply 125 by your own cost per qualified applicant to get the monthly media budget. If that lands far from what you were planning to spend, the cause is usually one of the four conversion rates being worse than assumed, not the budget.

Do not borrow last year's numbers. The euro area job vacancy rate was 2.1 percent in the second quarter of 2026, down from 2.3 percent in the first quarter and 2.2 percent a year earlier. Fewer open roles competing for the same feed inventory moves auction prices, so the only cost per qualified applicant worth planning against is your own, measured recently.

Frequently asked questions

Is there a minimum budget for job ads on Meta?
There is a technical account minimum, which varies by billing event and currency, but it is almost never the number that matters. The practical minimum is whatever produces a steady weekly flow of your chosen conversion event in each ad set, because below that the ad set stays in the state Meta labels as not generating enough results.

Should I run one campaign per role or one per region?
Per region, in most cases. One ad set per role at frontline volumes usually splits the budget too thin to stabilize, and roles in the same region compete for the same candidates anyway.

How long before I can judge the result?
Give it at least a full week after the last significant edit, because the conversion count that decides the learning state restarts at every significant edit.

Does a bigger budget lower cost per applicant?
Not directly. A budget large enough to leave the learning phase usually produces a more stable and often lower cost per result, but past that point extra spend raises costs as you reach less responsive audiences.

The short version

• Work backwards from hires through four conversion rates, and buy the unit closest to a hire that still occurs often enough to measure.

• Define the unit in the contract. A qualified applicant should mean completed flow, passed knock out questions, met job criteria and delivered into the ATS.

• Fewer, better funded ad sets beat many thin ones, because an ad set that cannot exit the learning phase never gets cheap.

• Start on lowest cost, read the real cost per result for a week, then set a cost cap just above it.

• Let the campaign distribute budget across regions unless you have to guarantee spend per client or per location.

• Expect the employment declaration to make audiences blunter, and judge performance on cost per qualified applicant rather than cost per click.