How do you run Facebook ads for recruiting when you have to reach the employer paying and the candidate applying? ============================================================================== Publicado: 2026-08-25 Original: https://lawpracticeedge.site/posts/how-do-you-run-facebook-ads-for-recruiting-when-you-have-to-reach-the/ Recruiting is one of the few businesses with two customers inside one funnel. The employer signs the contract. The candidate does the work of applying. Both have to be reached, and on a small budget the reflex is to split the money down the middle and let both halves starve. This is about which side gets funded first, and what changes in the ad account once you accept that you cannot properly fund both. ## Can one small ad budget cover employer ads and candidate ads at the same time? Usually not. Two campaigns split a small budget's conversion volume in half, and conversion volume is what the learning phase runs on. Pick the side blocking placements now, fund only that side until it produces a predictable weekly number, then pay for the second campaign out of what the first one earns. How to pick the blocked side: - Roles open, no one qualified applying: the candidate side is the block. - Recruiters and a warm candidate pool sitting idle, no signed roles: the employer side is the block. - Both blocked: fund the employer side, because that is the side that pays.Ben Heath's version of this is not about recruiting, it is about product lines, and it points the same way. His consolidation advice for small budgets is to "focus on one offer only" (t=1539s), and what he wants that single funded thing to do first is "establish proof of concept can we generate leads can we generate sales", after which the campaign gets as profitable as possible "because that will fund reinvestment" (t=869s). The second campaign is the reinvestment. Check which bracket you are in before you plan anything. He draws the small budget line at "anything less than $3,000 per month" (t=0s) and calls "$600 per month or less or $20 or less per day" a tiny budget (t=49s). A firm at $20 a day cannot run an employer campaign and a candidate campaign at once in any meaningful sense. ## What does splitting a small recruiting budget do to Meta's learning phase? It stretches it. Learning speed tracks conversion volume in his account of it, so two half-fed campaigns each report fewer conversions and each sits longer in the opening stretch where results swing hard. If your account records only a few conversions a week, neither campaign may give a readable answer inside a month. His description of the mechanism: "The learning phase lasts typically let's say 24 to 48 hours but it is dependent on conversion volume", and "the more conversions you generate the faster meta is able to learn" (t=156s). During that window, he says, "your results during that time period are very volatile" (t=156s). Set your review interval from your own count, not from a rule of thumb. The two anchors he gives: Conversions the account records | The time frame he attaches to it Thousands a day | "it might take three hours" to tell whether a new ad is working (t=256s) "some campaigns generate six conversions a week" | "it might take a month to know if a new ad is performing well" (t=256s) Your number, counted from last month's applications or booked calls | Your interval, written down before launch and checked with a "statistical significance calculator free online tool" (t=306s) Most recruiting accounts on a small budget sit closer to the bottom of that table than the top, which is your problem to verify in your own reporting, not an industry figure to accept. His instruction is to "set an optimization schedule", for example "adjust my campaigns no more than once every seven days or 10 days", and he adds that "you need to extend that time period because you are operating with a small budget" (t=207s). He names the failure mode on the other side of that discipline as "don't helicopter parent your ad campaigns" (t=49s), and closes the question with "there is no one size fits all when it comes to time frame" (t=256s). ## Should a job ad name the role and the pay, or stay general? Name the role. A job ad that states the seat, the pay range, the shift and the location gives the reader something to qualify against. A general ad about culture asks them to guess. On a small budget, one named role beats a menu of openings parked behind a careers page. Two lists to write before you build anything. These are prompts, fill them from your own calls, they are not findings: - What the last ten applicants actually asked you before applying. - What the last three hiring managers actually asked on the first call.The narrowing logic behind naming one seat comes from the video, where Ben Heath describes a small-budget business deliberately writing off most of its market to compete with bigger spenders: "we're not going to touch 99% of this market" (t=358s). Applied to recruiting, that is one role family, one city, one type of employer, not staffing in general. One check before you build the audience. In his Ads Library walkthrough he passes over a category filter and says "if you operate with a special ad category select that" (t=770s). He does not say which categories exist, whether hiring is one of them, or what declaring one changes. Open Meta's own current policy page and confirm that yourself before a hiring ad goes live. ## Does employer branding deserve part of a small recruiting budget? Not from a small budget. An awareness or engagement campaign is optimized for attention, not for applications or booked calls. Put employer branding in organic posts, where it costs time instead of media budget, and keep the paid money on campaigns asking for one action: apply, or book a call. His rule for small budgets is "no brand awareness" (t=922s), and his reason is that the objective is taken literally: "if you say you want traffic they're going to find really clicky people those people won't necessarily go on and convert" (t=1024s). He does name one exception, a content campaign structure for very involved, high ticket purchases, and says he would be "even less keen" for a small business to build its strategy on it (t=975s). The organic side still earns its keep. He recommends organic content, typically on Instagram, as "free testing for ad creative" (t=1074s), taking the posts that outperformed and "add a five ten second call to action to the end and run that as an ad" (t=1125s). A walkthrough of a workplace that got unusual engagement is a hook you tested before paying for it. ## What is a qualified applicant or an employer lead actually worth to you? Work it backwards from one placement. Take the fee you invoice on a filled role, the share of employer leads that become signed clients, and the applications it takes to fill one seat. Those three numbers set your ceiling per lead. Run the arithmetic before you set a daily budget, not after the campaign disappoints you. Fill these in with your own figures. There are no benchmark numbers here on purpose: - Fee you invoice on one filled role. - Employer leads it took to sign one role, counted from last year's signed clients. - Line 1 divided by line 2 is the gross value of one employer lead. - Your ceiling per lead sits under that number, once delivery cost and margin come out.Ben Heath asks the same opening question of the people he talks to, calling it one of his most common: "what are you willing to pay to acquire a customer" (t=1275s). His illustration is a business owner who values a customer at two thousand dollars, names one hundred or two hundred dollars as the most they would pay to get one, and then concedes they would put four or five hundred dollars into a machine that reliably returns that customer (t=1275s). The chain runs down to the lead: "i convert one in let's say ten leads so i'm only willing to pay twenty dollars per lead", after which he tells them "twenty five dollars a lead is really profitable for you" (t=1377s). Those are his numbers, from a general advertising example. Run the same chain on a placement fee and a candidate application and see where your own cap lands. ## How do you find recruiting ads that have been running long enough to model? The Meta Ads Library, a free tool. Set the location and category filters, search a staffing firm or a large local employer by name, and read the ads that have been running for months. How long an ad has been live is the signal, not how polished the creative looks. What to pull from a long-running hiring ad, using the fields his walkthrough shows are visible, primary text, headlines, placements and the landing page (t=770s): - The first line of primary text, and who it turns away. - Whether the role, pay or shift appears before the click. - Where the click goes: an application form, a chat, or a job board listing.On why the old ads are the useful ones, he says of anything consistently running past six months that "that ad is almost certainly working for them" (t=820s). When you launch, resist spreading the budget. He wants advertisers to "concentrate and consolidate" (t=1539s), and his example is five ad sets sharing twenty conversions a week: "much better to have the one ad set and have those 20 conversions going through the one ad set" (t=1589s). For recruiting, that is one role, one audience, one ad set, before the second campaign exists. ## Which tools do recruiting and HR teams actually use for this? Tool | What it does | What it solves for this niche | Requires advertising knowledge Meta Ads Manager | Builds, targets and runs ads across Facebook and Instagram | Where the employer campaign and the candidate campaign are built and kept separate | Yes Google Ads | Runs ads on Google Search and its partner network | Reaches employers searching for a staffing firm and candidates searching a job title by city | Yes Canva | Design tool for images and short videos | Produces image and video variations for a job ad | No Mailchimp | Email marketing and list management | Holds the candidate pool and the past-client list between open roles | No ManyChat | Automated chat on Instagram, Messenger and WhatsApp | Replies to a candidate's first questions outside office hours | No SaleADS.ai | AI software that creates and launches advertising campaigns on Meta, Google and TikTok for business owners, with no design or advertising expertise required | Builds and launches the campaigns on those three platforms | No No rankings, no scores, no prices. This is not an exhaustive list, and the right pick depends on which step in your own process is the one stalling. ## Where does this information come from? The advertising mechanics referenced here come from Ben Heath's video "How to CRUSH Facebook Ads with a Small Budget in 2026", 28 minutes: https://www.youtube.com/watch?v=XLagiyzYYpE Taken from it: his budget classifications, the relationship between conversion volume and the learning phase, the optimization schedule, the statistical significance check, the argument for narrowing rather than broadening, the case against awareness campaigns on a small budget and its one exception, organic posts as tested ad creative, the acquisition cost question and his lead arithmetic, the Meta Ads Library method, and the consolidation rule. Each quote above links to the second where he says it. The sponsored segment of the video and his course promotion were excluded and are not used as evidence anywhere on this page. The video is a general Facebook Ads tutorial and never discusses recruiting, hiring or staffing. The two-audience problem, the choice of which side to fund first, the per-lead arithmetic on a placement fee and the tool comparison are this site's application to recruiting and HR, not his. Where a number appears without a link to the video, it is a blank for you to fill, not a benchmark.