This one is short, because the lesson is short, and we've seen it cost people real money.
When you launch a campaign, Meta asks what to optimize for. Most people pick the easiest event to fire: a lead form, a page view, a "thank you" page. Meta then does exactly what it's told. It finds people who fill in forms. Not people who book calls, not people who buy. People who fill in forms.
The gap
On one account we inherited, the dashboard showed a cost per lead anyone would be proud of. The calendar told a different story. Most of those leads never booked, and the ones who did rarely showed. The campaign was winning at the wrong game.
We moved the conversion event to the moment a call was actually scheduled, confirmed by the calendar tool rather than a form. Then we checked it against a tracker that matched bookings back to the ad that produced them, because Meta's own attribution is generous with itself.
What changed
Cost per "lead" went up. Cost per booked call went down. That's the trade, and it's the right one. A rule we now apply on every account: any ad producing booked calls under the agreed cost doesn't get touched without a good reason, however it looks on the surface metrics.
Two things to check this week
First, open your ads manager and look at which event your campaigns optimize on. If it's a form or a page view and you sell through calls, that's the problem. Second, compare the dashboard's booked-call count to your actual calendar for the same week. If the numbers disagree, the tracking needs fixing before the ads do.
