How ATT works
App Tracking Transparency is a rule Apple applies to every app in its store. If an app wants to follow what a person does in other companies’ apps and websites — the thing Apple calls tracking — it must first show a standard system prompt and receive a yes. The wording of the prompt is Apple’s, not the app’s, and the answer is stored per app on the device.
What a yes actually unlocks is the device’s advertising identifier, plus permission to combine the person’s activity with data gathered elsewhere. A no does not stop the app working, and it does not stop advertising being shown. It stops the joining up.
Two limits are worth knowing. ATT governs apps, not websites, so a browser visit is a different matter governed by cookie rules and consent. And it governs cross-company tracking, so an app can still measure what people do inside its own service.
Why ATT matters to advertisers
Advertising platforms that owned large apps built their measurement on exactly the joining-up ATT restricts. When a person declines, an ad seen on their phone cannot be reliably connected to a purchase made later somewhere else, so it is not reported in the ordinary way.
That is the root of nearly every measurement complaint in Meta advertising today: shorter attribution windows, delayed reporting, missing demographic breakdowns, and a permanent gap between platform figures and a business’s own sales records. It also shrinks certain audiences, because the pool of people a platform can identify with confidence is smaller than it was.
Nepal’s audiences are heavily mobile and skew towards Android, so the practical bite here varies more by client than in markets where iPhones dominate. A clinic advertising to affluent Kathmandu customers will feel it more than a shop selling to a broad national audience.
Where the confusion lies
Many people take ATT to mean tracking has been banned outright. It has not. It made cross-company tracking opt-in on Apple devices, and left everything else — analytics inside your own site, consented cookies, your own customer records — untouched.
A second misunderstanding is treating ATT and cookie consent banners as the same thing. They cover different surfaces and different laws, and complying with one does nothing for the other.
The third is expecting the effect to fade. It has not faded, and platform policy has kept moving in the same direction.
What to do about it
Build measurement that does not depend on being able to follow someone across companies. Collect customer details lawfully at the point of enquiry or purchase and use them to match events properly. Send conversions from your server as well as the browser, which is the practical case for pairing the pixel with the Conversions API.
Then set expectations. Tell clients before a campaign starts that platform-reported results will sit below their own order count, agree in advance which number the decision will be made on, and judge performance over a period long enough for delayed reporting to arrive.