The three places most Meta accounts leak money

Most Meta accounts I audit are burning money in the same three predictable places. Not exotic problems, not anything that needs a secret tactic. Just three leaks that quietly add up, and that almost nobody checks until the results stop making sense.

1. The account is optimizing on the wrong event

This is the big one. If your pixel is firing a conversion on a form-start, an add-to-cart, or a page view instead of a real lead or a real purchase, Meta will happily go find you more of the cheap, shallow action you told it to value. The cost per result looks great. The revenue does not move. You are paying for motion, not outcomes.

The fix is not clever. It is confirming, event by event, that the thing Meta counts as a conversion is the thing that actually makes you money. On more than one account, fixing this single setting has done more than any creative change.

2. The creative stopped being tested months ago

Winning ads fatigue. The audience has seen them, the click-through rate slides, and the cost per result creeps up so gradually that nobody notices until the month is bad. If your account has not had a fresh creative angle in rotation for weeks, that slow drift is probably already happening.

A weekly testing rhythm fixes it. Not a full relaunch every week, just one new angle in, one tired one out, so the account always has something fresh feeding it.

3. Nobody is watching cost per result as spend grows

Scaling is where accounts break. You raise the budget, the algorithm reaches into less qualified audiences, and the cost per result drifts up until the extra spend stops paying for itself. The number that matters is not how much you are spending. It is whether the cost per result holds as you spend more.

The discipline is boring and it works: scale in steps, and make each increase hold the number before you take the next one.

The quick self-check

Pull your account open and ask three questions. Is the conversion event a real lead or sale? Has the creative changed in the last month? Did the cost per result hold the last time you scaled? If any answer is no, that is where your budget is going.

If you want a second pair of eyes on it, send me your worst-performing campaign and I will tell you what is wrong on the first call.

Frequently asked questions

How do I check which event my account is optimising for?

Open each ad set rather than the campaign, and read the performance goal and conversion event fields. Check every ad set individually, because the setting lives at ad set level and one can easily differ from the rest after somebody duplicated a campaign.

Will fixing the conversion event lower my Meta ads wasted spend?

It will cut the waste and usually raise your reported cost per lead at the same time. The system stops chasing a cheap shallow action and starts looking for people who genuinely enquire. Fewer, better, more expensive leads is the trade you want, so judge it on qualified enquiries.

How often should creative be refreshed?

On a rhythm rather than when performance drops. Waiting for fatigue to appear in the numbers means paying for a week of decline before you react. Introducing one new angle weekly and retiring the weakest keeps something fresh arriving before the current winner burns out.

What counts as a genuinely new creative angle?

A different reason to care, not a different colour. The same offer with a new headline is a refresh. The same offer framed for a worried customer instead of a bargain hunter is a new angle, because it reaches a different group of people entirely.

Is a rising cost per result always a leak?

No. If you raised the budget, some rise is the auction working as designed, because the cheapest attention gets bought first. It becomes a leak when the figure crosses what a lead is worth to you, which is profit per sale multiplied by close rate.

Are there other places Meta accounts waste money?

Plenty, but these three account for most of it in the accounts I open. Missing exclusions for existing customers, geography that does not match where you actually trade, and untracked phone enquiries are the next three worth checking once these are handled.

How quickly will I see a difference after fixing these?

The conversion event change takes a week or two to settle, because you are asking the system to unlearn what it was previously optimising towards. Creative changes read faster. Expect a wobble before an improvement, and do not reverse anything in the first week.

Can I check all three myself?

Yes, and the self-check in this article takes about ten minutes. What takes experience is deciding what to do with what you find, particularly whether to accept a higher reported cost per lead in exchange for enquiries your sales team can actually use.

1. The account is optimizing on the wrong event

2. The creative stopped being tested months ago

3. Nobody is watching cost per result as spend grows

The quick self-check

Frequently asked questions

Send me your worst-performing campaign.