Glossary
What is an install referrer?
The install referrer is data passed from an app store to a newly installed app, identifying which ad click, campaign or referral source drove the download.
How it works
A campaign link to the Play Store carries a referrer parameter. The store retains it through the install, and once the app first launches it can query the store for that value.
The result is deterministic attribution that does not require matching device identifiers at two points in time. The store itself carried the campaign information across the gap, which is a fundamentally more reliable mechanism than probabilistic or identifier-based matching.
Why it matters more than it used to
As device identifiers became unavailable or consent-gated, mechanisms that do not depend on them gained importance disproportionately.
The install referrer is one of the few deterministic signals that survived intact on Android, because it is passed by the store rather than derived from tracking the user. Newer versions of the API also carry timestamps that support click to install time analysis, which is a significant fraud detection input — the mechanism that ended most click injection.
Its limits
- Android only. iOS has no equivalent; SKAdNetwork fills that role differently.
- Play Store installs only. Sideloaded and alternative-store installs carry nothing.
- Single value. It records one referrer, so it cannot describe a multi-touch path.
- Correct implementation required. The app must query the API on first launch; a missed integration loses the data permanently for that install.
Common questions
Is the install referrer affected by privacy changes?
Far less than identifier-based attribution, because it carries campaign information rather than user identity. It has remained one of the more stable signals on Android.
Can the install referrer be spoofed?
It comes from the store rather than from the network claiming credit, which makes it considerably harder to fabricate than a click record — one reason it is trusted for fraud analysis.
More in growth and engagement
Acquiring users, and the behaviour that says whether they stayed.