How it works
The CMP shows a notice listing the purposes and vendors involved, records what the user accepts or refuses and exposes the result to other code in the app. The result can take the form of a standard consent string under the IAB Europe Transparency and Consent Framework, or of signals such as those used by Google's Consent Mode. Advertising and measurement SDKs read these signals and are expected to adjust, for example by not sharing data with ad partners for a user who refused. The user must be able to reopen the screen and change their mind.
Why it matters for an app team
In the European Economic Area and the UK, storing or reading identifiers on a device for advertising generally requires consent. Large ad platforms also expect consent signals for European users before they use conversion data for personalised advertising, so a missing signal can mean lost campaign measurement as well as legal risk.
In practice
Example with made-up numbers: of 8,000 new users in the EEA, 5,600 accept advertising purposes and 2,400 refuse. Conversions from the 5,600 can be shared with ad partners. For the 2,400 they cannot be used for ad personalisation.
Common confusion
A CMP is not Apple's App Tracking Transparency prompt. ATT is a system permission for the IDFA; a CMP records legal consent, and an iOS app serving European users typically needs both. A CMP also only records choices. Whether each SDK respects them is for the app team to check. This entry is a general explanation, not legal advice.