The figure that worries people
Of my 26 ads from 2023 to 2025, 24 had reliable visit tracking: together they had 24,428 visits. 5.6% of visitors applied, which means 94.4% left without leaving anything.
That sounds huge, and it is less so than it seems. An ad reaches people who had not asked for anything. Many take a look and close it: that is the nature of the medium. The figure does not say the page is bad; it says how many curious people you need to find the ones who are ready.
This figure comes from my history, measured by Facebook's tracker on the page. I cannot repeat it for 2026, and I explain why below.
Where candidates drop off, step by step
| Step | What I measure | My figure | Period |
|---|---|---|---|
| The click | people who see the ad and click the link | 1 in 125 (0.8%) | 2023-2025 |
| Loading | link clicks that end with a page displayed | 70% | 2023-2025 |
| The decision | visitors who apply | 5.6% | 2023-2025 |
| The form | people who start and go all the way | 34% (from 31% to 38%) | 2026, 5 clients |
Two losses strike me:
Three clicks in ten never reach the page. The person clicks, the page takes a while to open in the app, and they go back. Part of this loss also comes from the tracker itself, which does not see everyone. I do not know the split, and I would rather say so.
Two people in three give up while filling in the form. Of 1,007 people who started applying in 2026 with five clients, 340 went all the way. The figure is stable from one client to another, from 31% to 38%. It is an approximation: I count unique visitors who started the form and the applications received in the same window.
What I do to limit the damage
These are design choices, not measured results:
- A light page, built to open quickly inside Facebook and Instagram on a phone.
- Few fields: people can apply with or without a CV.
- Short questions that can be answered by speaking, because typing a long text on a phone drives people away.
I have not tested the effect of each of these choices separately. What I measure is that about a third of the people who start finish. That is the figure to raise.
And the ones who go all the way, what are they worth?
That is the right question, and the answer depends on the employer as much as on the advertising. Here is what three of my clients did with the applications they received, according to the statuses they set in the tool:
| Client | Applications received | Contacted, shortlisted or accepted | Rejected |
|---|---|---|---|
| Servagroupe | 90 | 54 | 32 |
| Angers Loire Métropole | 47 | 17 | 19 |
| Édifice Industrie | 128 | 10 | 108 |
From 8% to 60%: the gap mainly says what each employer was looking for and how strictly they sorted. A client who rejects most of their applications may simply have a very demanding job. So I do not take an average, and I do not promise you a contact rate.
Nor do I give you the number of hires: “accepted” does not mean “hired”, and employers do not always tell me who started.
In summary
- 2023-2025: 94.4% of visitors leave without applying. That is normal for an ad, and it could not be measured again in 2026.
- Only 70% of link clicks turn into a visit: page speed matters, and tracking also loses a share.
- 2026: about a third of the people who start the form send it, from 31% to 38% depending on the client.
- What happens to applications depends on the employer: from 8% to 60% contacted at three clients.
