john...@chromium.org, cshar...@chromium.org, lin...@chromium.org
https://github.com/WICG/attribution-reporting-api/blob/main/app_to_web.md
https://wicg.github.io/attribution-reporting-api/#cross-app-and-web (a rough shell at this point)
Currently, the Attribution Reporting API supports attributing events within a single browser instance. This proposal expands the scope of attribution to allow attributing conversions that happen on the web to events that happen off the browser, within other applications such as mobile applications.
The proposal here takes advantage of OS-level support for attribution. In particular, it gives the developer an option to allow events on the mobile web to be joinable with events in Android’s Privacy Sandbox, although support for other platforms could also be implemented in the future.
The experiment will be on Android T devices only.
Blink componentInternals>AttributionReporting
https://github.com/w3ctag/design-reviews/issues/724 (Attribution Reporting API)
Pending
There are several other different attribution measurement proposals from a variety of browser vendors and companies, each offering different forms of privacy and utility:
Safari has proposed and implemented Private Click Measurement.
Mozilla has proposed Interoperable Private Attribution.
Microsoft has proposed a few ideas (MaskedLARK, Bucketization).
The following signals are for the Attribution Reporting API, we haven’t requested signals separately for the cross app and web extension yet.
Gecko: No official position
WebKit: No official position
Web developers: TBD
N/A
A successful API flow involves enabling Measurement API on Android and registering events on the web to pass to Android. Android is responsible for the actual attribution and reporting.
N/A
Security
Security considerations are detailed here.
Privacy
Privacy considerations are detailed here.
For experimentation with the new extension of the Attribution Reporting API, we hope to see that the measurement data made available through the API provides useful ad conversion data cross app and web.
Experiment Configuration
The origin trial for this experiment will be separate from existing Privacy Sandbox APIs to reduce risks for other APIs in case of bugs in this origin trial.
The new origin trial will support third-party origin trial configurations. To ensure that developers can run this experiment without concern for crash/breakage, this origin trial will be available for a subset of users by default. Initially, the Cross App And Web Attribution Measurement API will only be enabled on 50% Canary/Dev Chrome channels. As we gain confidence that the API is working properly by monitoring feature usage, we will enable the API for the Beta Chrome channel and eventually the Stable Chrome channel.
The Attribution Reporting API will be fully enabled (app-to-web, web-to-app, web-to-web) with the new origin trial token.
None.
The Attribution Reporting API utilizes DevTools and an internal page (chrome://attribution-internals) to facilitate testing and integration. Debug reports are supported (and when configured in a third-party context, require third-party cookies to be available).
The debugging information for OS registrations will be displayed in DevTools and in chrome://attribution-internals as well. Android Measurement is also implementing a similar debugging reports framework to facilitate cross app and web testing.
No, only on Android for this experiment.
No, web platform tests are not supported for Android.
The API can be enabled using chrome://flags/#privacy-sandbox-ads-apis.
No.
https://launch.corp.google.com/launch/4238495
We hope to start the Origin Trial sometime during M114 beta (M114 to M119 inclusive).
https://chromestatus.com/feature/4994430156668928
Attribution Reporting API Intent to experiment:
https://groups.google.com/a/chromium.org/g/blink-dev/c/jEnNpideO1Y/m/NXiiV_nnDwAJ