Why your Facebook ads are not working, and where to start looking

You launched an ad, spent a little money, and nothing happens. Or something happens, but not what should. The first reaction is almost always to touch the settings: change the targeting, raise the budget, relaunch.

That is rarely where the problem sits.

The reflex that costs money

Settings are what you can see, so they are what you change. Every change also restarts part of the campaign's learning, which blurs the reading further.

When I look at an account that is not performing, I do not start with the ad. I start with what happens before and after it.

Five places to look, in order

The order matters. The lower a problem sits in this list, the more serious it is, and the less likely anyone is to think of looking for it.

1. The offer. Does a person seeing your ad understand what they get, for whom, and why from you? If the offer is vague, the ad only lets more people see the vagueness. No setting fixes that.

2. What happens after the click. A slow page, a message that does not look like the ad, an irritating form: the person arrives, hesitates, leaves. The ad did its job, the contact was lost just after.

3. The creative. Does the video stop the scroll, and does it deliver what its opening promises? It is the lever that weighs most on what Meta decides to show, and to whom.

4. The audience. Rarely the root cause, and often the first place people start tinkering.

5. The settings. They amplify what works. They repair nothing.

Most advertisers work through this list backwards, bottom to top, because that is the order in which the screens present themselves to them.

What that changes in practice

A campaign whose problem is the offer does not improve by changing the audience. A campaign whose problem is the landing page does not improve with a new video. Every misplaced fix costs budget and tells you nothing more about the real cause.

Before touching anything, one question: is the bottom of this list solid?

Judging too early, the other trap

An ad that brings in nothing after a week is not necessarily a bad ad. The first days are for learning, and on a long purchase the first sales arrive well after the first contacts.

On the kitchen case published on this site, the first month signs no sale at all, even though quote requests arrive from the start. The decision cycle is six to ten weeks. A kitchen fitter who judges the campaign at thirty days concludes it does not work, and is wrong.

Conversely, some campaigns spend for several weeks with no result because the problem is real. The difficulty is knowing which, and it is not readable from the dashboard alone.

What makes diagnosis hard

A single figure says almost nothing. A high cost per contact can come from the creative, the landing page or the offer, and only a look at the whole journey settles it.

You also need reliable data. If tracking is incomplete, Meta sees fewer sales than there are, and the campaign looks worse than it is. A pixel alone is no longer enough for that reason.

Finally, you have to be able to compare what Meta reports with what you take in. Without that cross-check, you optimise without knowing whether the result exists.

When to ask for an outside view

If you have spent for several weeks with no result, if you do not know which of these five places is the problem, or if your figures do not match what you take in, an outside view costs less than another month of budget.

---

Further reading: Why your ads cost more and more and Why Facebook quote requests have a bad reputation. If the account itself is blocked: restricted ad account, what to do. The full list: auditing an ad account.

Your ad is not bringing anything in? Book a 30-minute call. We look at your account together, and we look in the right place.