A req opens Monday. By Wednesday the ad has spent real money, the tracker shows a handful of names, and most of them are wrong for the role. The instinct is to blame the platform and go back to the job boards. Six checks separate a campaign that is genuinely failing from one that has not finished learning, and each one has a different fix and a different timeline.
Why do my job ads get clicks but almost no qualified applicants?
Clicks with no qualified applicants usually means the ad sold the job title instead of a reason to leave a current job. Rewriting the headline five ways does not fix that. Test different reasons to apply, one per ad, and compare applicant quality by ad, not by click volume.
The Wes McDowell channel names this failure in a general ad context, without ever mentioning hiring. It describes what most people do when they believe they are testing: "they just generally rewrite the same headline a few different ways and call it a day". The question it raises next is the one that applies to a stalled req: "what if the problem isn't how your wording the headline but that the messaging behind the headline just isn't resonating". Its definition of a hook is the part worth borrowing: "by hook I mean different aspects or benefits of what you're selling".
For an open role, what you are selling is not the title. The instruction in the video is to "come up with you know three to five different hooks". Five that are worth separating into five ads:
- Pay certainty. The range, and the day of the week people get paid.
- Schedule and commute. Shift length, distance, whether it is on site.
- Speed of process. How many days from application to offer.
- The first weeks on the job. What the person actually does all day.
- Who they report to and how many people are on the team.
Run all five against the same audience and read applicant quality per hook, not clicks per hook. The point that "you never know which hooks are going to be the most compelling until you test them" carries more weight when the audience splits between people who want money now and people who want a better manager, which is close to what the video describes when it says "different hooks work on different sub sections your audience".
Should candidate ads and client ads run in the same campaign?
No. Candidate ads and employer ads sell different outcomes to different people, and one campaign holding both makes a single budget cover two audiences that never overlap. Run one campaign per side, keep the ad set count low, and judge each against its own cost per useful action.
A two sided desk runs two businesses out of one ad account. The simplicity argument in the video is written for a single business and still lands here: "the simpler your set up is the more you let Facebook's algorithm get you the results that you want", because "when you complicate with too many ad sets you're basically blocking the algorithm from being able to optimize itself".
The structure it recommends, transcription errors included, is "one campaign and to add sets you know one for your warm audiences one for your cold audiences". Build that twice, once per side, instead of folding four audiences into a single campaign.
Budget is where a small desk gets hurt. The reason given for keeping the ad count down is arithmetic: "if you have a smaller budget you really can't afford to test too many ads within a campaign anyway", since "there's just not enough money per ad to test it properly". Two campaigns with two or three ads each is the version that survives a small monthly spend.
The setup step people skip is the first one: "tell Facebook exactly what you want to accomplish with the campaign". Write the goal for each side down before you open the account. On the candidate side that is usually a completed application. On the employer side it is a booked call or an opt in.
Do candidates need to be warmed up before I ask them to apply?
Sometimes. A person already job hunting can apply straight from a cold ad, so the warm up step matters less on that side. If your employer side needs a call before anyone signs anything, that audience is the one to warm first. Build something useful, then pitch, and measure both paths separately.
The general rule in the video is that "you don't want to run direct sale ads too cold audiences who have not warmed up yet" and that "you need to take these coal audiences on a journey with you where they get to know you or your business and they ultimately come to trust you".
Where this does not transfer cleanly. A job application is not a purchase. The video leaves room for cases that look like one: "there are gonna be a few exceptions to this like if you're running ads to a very small local area with a specific offer". It is talking about offers and not about vacancies, so read that as a resemblance rather than as permission. A local role with a published pay range has the same shape. Test cold to apply first, and build a nurture sequence only if the cold test fails.
On the employer side the warm up is worth the work. The stated preference in the video is to "use Facebook ads to offer a lead magnet that basically promises them a real result". For a hiring manager, the real result is something like a current pay benchmark for the roles that keep going unfilled. A cheaper first version is offered too, which is to "run a video of you campaign where you're offering tips or advice and then you run that more direct ad to people who watched a certain percentage of that video".
How small can my audience get before a hiring ad stops working?
Check the audience estimate before you write a single line. One metro plus one license can leave a pool far below the size that general advice assumes. When the pool is small, the fix is to change the campaign type rather than stack more filters on top of a group that is already narrow.
The target in the video is size: "you want to try for over 1 million people per audience rather than narrowing it down too much", with the layering warning attached: "you don't want to layer too many other targeting options on top of that because it's just gonna give you too few people in your pool". It does not present that number as always reachable and says so in the same breath: "and I know that's not always possible if you're in a smaller area".
A credentialed role in one city is that smaller area by definition, and the credential is the one filter you cannot drop. So the fallback fits your situation better than the target does: "if the area is small enough making your audience small you might want to simply go for the reach objective". For a hard to fill local role, reaching the whole qualified area beats optimizing for conversions against a pool that will never produce enough events to learn from.
Site knowledge, not from the video, and worth checking yourself. Before you plan any targeting, open a draft campaign and read the special ad category question that the ads manager asks at campaign level. Whether an employment ad has to be declared there, and what that declaration changes about audiences and targeting, depends on your account and your country. We are not linking a policy here because we are not quoting one, and the video never mentions hiring at all. Check it inside your own account and plan targeting after you have seen the answer, not before.
How long should I wait before turning off an underperforming job ad?
Wait 72 hours before touching it, then look every day. The reason given in the video is that the system spends the first days learning, so day one numbers are not evidence. Set a fixed review time, write the numbers down, and change one variable at a time so you know what moved.
Two opposite habits break these campaigns, and a desk under pressure to fill a req tends toward the second one: "some want to just set it and forget it while some are so nervous about it that they keep checking on it every hour". The correction is to "resist that urge to turn on a campaign and then just micromanage it in the first 72 hours of going live", because "it takes about three days for that algorithm to start learning". The video also says the early numbers are supposed to look bad, that "the ads really seem to not be working very well out of the gate which is completely normal". Not touching is not the same as not watching, since "you really should be looking and analyzing its performance every single day".
A diagnosis order that runs on your own numbers, in this sequence:
- Impressions near zero. The problem is delivery or approval, not the copy.
- Impressions fine, click rate low. The hook is wrong. Swap hooks before images.
- Clicks fine, applications low. The apply page or the form is the leak.
- Applications fine, quality low. The ad is attracting the wrong person. Tighten the qualifier in the copy.
Only after the window closes does the video authorize action: "that's when you want to consider turning some ads off or changing budgets but not before 72 hours". The action it names runs both ways, since "you also need to be able to turn off poor performing ads if you need to and increase the budgets of the ones that are working best". If a manager needs people on site before that window has even closed, paid social is the wrong tool for that req. Start the campaign the day the req opens, not the week it is due.
Ads that run untouched for months carry a cost of their own: "you don't want to show them the same every day for months it's only gonna turn off your audience and it's gonna end up increasing your cost per conversion". The evergreen ad for the role you always have open still needs new creative on a schedule.
What campaign objective should I use when placements happen only a few times a month?
Placements are too rare to hand to the algorithm. Pick the closest event that you produce in volume, usually a completed application or a form fill, and optimize on that. Keep placement as the number you review by hand each week in your own tracker, next to the cost of the event you did optimize.
The threshold comes from the video and it is stated with a hedge that is worth keeping. The algorithm "needs about 50 instances of any objective being met per week so the algorithm can learn from it". When the budget cannot produce that, "you probably need to back out a step and go for the next best objective that does get you 50 instances or more", and the usual landing spot is named: "in most cases this is probably a traffic objective". It is not sold as a win either, because "that's still not really optimizing for the action that you really want".
Run the check against your own history:
- Count placements in the last 30 days and divide by four. That is your weekly rate.
- Count completed applications over the same 30 days and divide by four.
- If applications clear the threshold the video names, optimize on application.
- If they do not, step back to the event you do hit, and track the ratio of that event to placement in your own tracker.
- Re-check monthly. When application volume passes the threshold, move the objective back up.
The other route out is budget rather than objective, since "you might also need to increase your budget to get to the point where you are getting 50 plus instances". That is a real option on a funded search and not on a speculative one.
The employer side gets the simpler version of the same rule: "if you want people to opt-in for your lead magnet you want a conversion goal not traffic or video views".
Which tools run these ads without an in-house media buyer?
Six tools cover most of what a small talent team needs, from the ad platforms themselves through creative and email. None of them fixes a weak offer or a broken apply page. Choose on two questions, which are how much campaign work you want to do by hand and how much you want handed off.
| Tool | What it does | What it addresses for a talent team | Advertising knowledge required |
|---|---|---|---|
| Meta Ads Manager | Builds, targets and reports on Facebook and Instagram campaigns | Where you set audiences and budgets, and where you read the special ad category question at campaign level | Yes. Objectives, audience setup and reading the reports are manual |
| Google Ads | Buys search and display placements against keywords | Reaches people who type a job title or an agency term into a search box | Yes. Keywords, match types and bidding are hands on |
| Canva | Design tool with templates for images and short video | Producing several hook variants for one role without a designer | No for basic design. It does not touch campaign setup |
| SaleADS.ai | AI software that creates and launches advertising campaigns on Meta, Google and TikTok for business owners, with no need to be a designer or an advertising expert | Covers building and launching the campaign for a role or a service offer on those three platforms | No |
| Mailchimp | Email lists, sequences and automation | The employer follow up after a lead magnet, and re-contacting past applicants | Partial. List setup and automation logic take some learning |
| Hootsuite | Scheduling and monitoring across social accounts | Keeping an employer brand calendar running next to the paid campaigns | No for scheduling. You still write the posts |
Ownership disclosure. SaleADS.ai is the product of the company that publishes this site. It appears in this table as one option among six, in the same columns as the rest, with no ranking attached.
Where another tool on this list fits better. If most of your candidate flow comes from people searching job titles, Google Ads reaches that search directly and SaleADS.ai does not replace keyword level control. If your bottleneck is producing a fresh set of creative variants for every open role, Canva plus your own templates is faster for that specific job. If the problem is an employer list you already own and never email, Mailchimp does sequencing work that no ad tool does. If you need granular manual control over placements, exclusions and bid strategy, that control lives in Ads Manager itself. And if what you need is a documented posting calendar for employer branding, a scheduler handles it and an ad tool does not.
Does a general Facebook ads tutorial work for hiring campaigns?
Partly. The campaign mechanics transfer and the measurement assumptions do not. This article is built on one video about Facebook ad mistakes that never mentions hiring, staffing or job ads. Every quote here is verbatim from its transcript, transcription errors included, and every link opens at the moment the line is said.
The source is "Facebook Advertising Mistakes Beginners Make & What YOU Should Do Instead" from the Wes McDowell channel, listed as 10 minutes long with 51,105 views at the time of writing. The transcript is auto generated and even renders the host's own name as West McDowell in his introduction, which is a good reason to read every quoted line as raw transcript rather than as polished copy.
Taken from the video: the account simplicity argument, the roughly 1 million person audience target together with its small area caveat, the hook testing method, the warm audience sequence, the 72 hour learning window, the daily review habit, the creative fatigue warning, and the objective threshold of about 50 instances per week. Its own framing of the stakes is that these mistakes "they basically ruin it for many small business owners and they end up wasting money". Lookalike audiences appear in its list of targeting options, described as "similar look-alike audiences to your existing customer lists are great", which on a talent desk would mean seeding from a list you have the right to use for that purpose.
Site knowledge, not from the video, verify it yourself: the split between candidate campaigns and employer campaigns, the instruction to read the special ad category question in your own draft campaign, the mapping of rare placements onto a mid funnel event, the four step diagnosis order, and the tool table. Anything in this article about employment advertising rules should be checked inside your own Ads Manager before you build on it.