How app promotion works
App promotion is the campaign objective used to send people to the App Store or Google Play and, more importantly, to find the ones who will do something once the app is open. Before it will run properly the app has to be registered in Meta’s developer settings, connected to the ad account, and wired to send events back — either through Meta’s own SDK or through a mobile measurement partner that sits between the app and every ad platform you use.
Once events are flowing, the ad set can optimise for the install itself or for a later in-app action: a registration, a first booking, a subscription, a purchase. Meta also offers an automated version of the objective that takes over audience, placement and creative combinations, which suits accounts with steady volume more than accounts that are still finding their footing.
Why app promotion matters
Installs are the easiest app number to buy and the least useful one to own. An app that is downloaded and never opened has cost money and returned nothing, so the question the objective exists to answer is not how cheaply you can be installed but how cheaply you can acquire someone who stays.
That is why the choice of optimisation event carries more weight here than almost anywhere else in Meta Ads. Optimising for installs finds people who like installing things. Optimising for a registration or a purchase finds people who use apps like yours, and it costs more per install for exactly the reason it should.
Common mistakes with app promotion
The first is launching before measurement works. If the SDK is not sending events, or the measurement partner is not passing them to Meta, delivery has nothing to learn from and the campaign quietly buys the cheapest installs available.
The second is comparing iOS and Android results as though they were the same measurement. Apple’s privacy framework means iOS results reach Meta in an aggregated, delayed and incomplete form, so aggregated event measurement reporting will look thinner than Android reporting for reasons that have nothing to do with campaign quality. The third is ignoring everything after the click. Store rating, screenshots, description, download size and how the app behaves on an older Android handset all decide whether a click becomes an install, and none of them are fixed by raising a bid — that is app store optimisation work, not ad work.
How to act on it
Get the plumbing right first. Confirm in Events Manager that installs and at least one deeper event are arriving, with the parameters you will later want to report on, and only then let a campaign spend.
Optimise for the deepest event that happens often enough for delivery to learn from. If purchases are rare, optimise for the step before them and watch what happens further down. Judge the campaign on the cost of a user who is still there weeks later, or who has paid, rather than on cost per install — and keep iOS and Android in separate reporting lines so neither one hides the other’s problems.