What do three years of Meta ads for recruitment actually deliver?

Across three full years, our ad account ran 195 recruitment campaigns on Meta, each one manually classified as such, for 65 clients with at least one such campaign in that window. The account as a whole, internal and other campaigns included, produced 28.7 million ad impressions, one ad impression being one time an ad was shown on a screen, and an estimated reach of 3.7 million accounts. That estimated reach figure is Meta's own estimate of distinct accounts, not a count of distinct people: someone with a personal and a business profile can be counted twice.

The same account recorded 444 thousand visits to the application form, where a visit means the form finished loading after someone clicked through, not that anyone filled it in. We count only ads that point to our own application form; 1.7% of the visits demonstrably went somewhere else and for 0.7% the destination can no longer be established, so neither is counted. Because the tracking pixel is not present on every form we use, this number is a lower bound rather than a precise total. The whole account ran on a campaign budget of around €400,000 spread across the period; internal and other campaigns are in these account figures, but not in the 195. These figures come from SocialFind's own Meta ad account and our own application forms, 10 August 2023 to 31 August 2026; the exact definitions sit on our measurement methodology page, and the campaign work itself is part of what we do in job marketing. The current value of each figure is on our figures page, which we update every month.

Over the same three years, our application forms recorded 24,316 completed submissions. That number counts form submissions, not unique people, since one candidate can submit more than once, and it is not attributed to the ads described above: a submission can come from an ad, from a shared link, from organic traffic to a vacancy page, or from a candidate who found the role some other way entirely. We report the two sets of numbers side by side because employers ask about both, not because one explains the other.

What we can say is how the funnel itself is built: how a click turns into a visit to the application form and, further along, into a submission is described on the application funnels page. Reading the two pages together gives a fuller picture than either number on its own.

Once someone reaches the form, their behavior looks a lot like ordinary phone use. In our own numbers, 95% of submissions arrive on a mobile, and close to half, 46%, come straight from the Instagram app, in the browser built into that app. Those shares are high because we advertise almost only on Facebook and Instagram and that audience is on a phone, so they describe this channel and not applicants in general. Filling in the form itself does not take long: the median time to complete it is 32 seconds, once we exclude outliers above two hours.

A large part of the submissions arrives outside office hours. 31% of submissions arrive after 6 pm, and 26% land in the weekend, with Sunday responsible for 14% on its own, as busy as a weekday. This too comes from our own forms, and we advertise almost only on Facebook and Instagram. None of this is unique to any one platform, but it does explain why the approach behind social recruitment leans on short, mobile-first forms rather than a long application built for a desktop screen.

Does a Meta ad guarantee applications?

No, and we do not present it that way. An ad impression means an ad was shown, a visit to the application form means the form loaded after a click, and a submission means someone finished the form. Each step loses people compared with the one before it, which is normal for any online channel. The numbers in this article describe scale and behavior over three years, not a promise about what a new campaign will produce, since every vacancy, audience and creative is different.

How is estimated reach different from ad impressions?

Ad impressions count how many times an ad appeared, so the same account can be counted many times if it saw the ad repeatedly. Estimated reach counts distinct accounts instead, though it is still Meta's own estimate rather than a verified count of people, since one person can hold more than one account. Over our three-year period the difference was large: 28.7 million ad impressions against an estimated reach of 3.7 million accounts, which shows how often the same account saw an ad more than once.

Why report visits to the application form instead of applications?

A visit to the application form is the most honest number we can measure close to the ad itself: the form finished loading after someone clicked through, on an ad that points to our own application form. We do not call it an application, because the pixel that counts it is not present on every form we use, so 444 thousand visits is a lower bound, not a finished count. Completed submissions are tracked separately, on the form itself, and are not linked back to a specific ad, which is why we keep the two numbers apart.

SocialFind builds and runs recruitment campaigns on Meta for employers who want to reach candidates where they already spend their time, from the first ad to the application form itself. If you would like articles like this in your inbox, sign up for our newsletter.