Definition
Privacy-preserving attribution is any approach to crediting installs and conversions to marketing campaigns that avoids following an individual across apps and websites. It grew out of platform and regulatory changes, notably App Tracking Transparency on iOS, which made cross-app tracking subject to explicit user permission.
Common Approaches
- Platform-generated attribution. The operating system produces the signal itself. Apple's SKAN and AdAttributionKit return limited, delayed postbacks.
- Thresholds and noise. Detail is released only when enough users share a campaign, an idea covered under crowd anonymity.
- Aggregate reporting. Results are summed across many users, as described under aggregate attribution, so no single user can be singled out.
- Consent-based measurement. Users who opt in can be measured at a finer level, while others are measured in aggregate.
- Data minimisation. Hashing personal data, collecting fewer fields and limiting retention.
What Changes for Marketers
- Fewer fields per install and longer delays before results settle.
- Conversion events must be chosen deliberately, since limited signals can only carry so much.
- Campaign-level reading replaces ad-level or user-level reading on affected traffic.
- Validation through incrementality testing and cohort trends matters more.
In Practice with Linkrunner
Linkrunner supports SKAN 4.0, consent signals and opt-in SHA-256 hashing of personal data in its SDK, which is off by default. For the wider picture, see Is Mobile Attribution Dead Under Privacy Changes? and Beyond Traditional Attribution: How to Track User Journeys in a Post-IDFA World.