How conversion location works
After you pick a campaign outcome, the ad set asks a narrower question: where should the thing you want to happen, happen? Depending on the objective the options include your website, your app, an instant form that opens inside Facebook or Instagram, a Messenger or WhatsApp conversation, a phone call, or your profile or shop on Meta’s own platforms.
That answer then shapes everything below it. It decides which performance goals and optimisation events Ads Manager will offer you, which tracking has to be working before the campaign is worth running, and what the reported result actually refers to. Under Meta’s outcome-driven objective framework, this is where most of the choices that used to live at campaign level now sit.
Why conversion location matters
Each location asks a different amount of effort from the buyer, and that effort is a filter. Sending someone to a website form asks them to leave the app, load a page and type; an instant form asks them to tap twice. The cheaper route produces more results and weaker ones, and the dearer route produces fewer and stronger ones. Neither is right in the abstract — it depends on whether your problem is too few enquiries or too many poor ones.
It also decides what your business has to be able to do. A messaging location is only as good as the person answering, which in Nepal often means WhatsApp or Viber replies within minutes rather than hours. A call location needs somebody picking up during the hours the ads deliver. Choosing a location your operation cannot service is a decision to pay for enquiries you will then lose.
Common mistakes with conversion location
The most common is choosing the website while the pixel is broken or the event never fires. Delivery is then optimising towards something it cannot see, and the campaign drifts towards cheap clicks by default.
The second is comparing cost per result across locations as though the results were the same thing. A message reply and a completed website checkout are not comparable, and putting them in one column of a report invites the wrong conclusion. The third is editing the location on a live ad set. That changes what delivery is looking for, restarts learning and destroys the comparison you were presumably trying to make.
How to act on it
Start from how your customers already behave rather than from what is easiest to set up. If people habitually message before they buy, a messaging location will beat a form. If the purchase is considered and the website does the persuading, send them to the website and accept the higher cost per action.
Confirm the plumbing for that specific location before spending: the event firing for a website, a working number for calls, a staffed inbox for messaging. Then, if you want to compare two routes, run them as separate campaigns and judge them on qualified outcomes rather than on the platform’s cost per result. The performance goal you can choose is constrained by this setting, so getting the location right is what makes the rest of the ad set worth configuring.