Get Inspired
Docs
Blog
Articles
The Privacy Sandbox
What is the Privacy Sandbox? Proposal lifecycle in the Privacy Sandbox Timeline Glossary and key concepts Privacy Sandbox events Privacy Sandbox demos Blog Privacy Sandbox videos
Chrome-facilitated testing Relevance and measurement unified origin trial Enroll your site API status and feature releases
Privacy Sandbox testing FAQs Privacy-related compliance FAQs
Feedback
Prepare for phasing out third-party cookies Related Website Sets Related Website Sets: developer guide How Chrome evolved the First-Party Sets proposal Cookies Having Independent Partitioned State (CHIPS) Storage Partitioning Fenced frames
Federated Credential Management API Federated Credential Management API: developer guide Federated Credential Management API updates
Shared Storage Shared Storage API walkthrough
IP Protection Privacy Budget Bounce tracking mitigations
User-Agent reduction User-Agent reduction snippets
Topics API overview Topic classification Developer guide Integration guide Experiment and participate Topics API demos Test topic inference in a colab Topics API latest updates
Protected Audience API Experiment and participate
Protected Audience API: developer guide Seller guide: run ad auctions Buyer guide: join interest groups and generate bids Protected Audience API auction reporting Sequential auction setup with header bidding and multi-seller Protected Audience auction Improve Protected Audience API auction latency Opt-out of the Protected Audience API Report on Protected Audience API auction results Troubleshoot the Protected Audience API
Status of pending Protected Audience API capabilities Protected Audience API walkthrough
Content selection with Shared Storage Creative rotation Creative selection by frequency A/B Testing Known customers
Maximize ad relevance
Attribution Reporting overview Attribution Reporting end-to-end Experiment and participate Why Chrome plans to ship the Attribution Reporting API
Introduction to debug reports Set up debug reports Debugging cookbook
Attribution Reporting API developer guide Get started with Attribution Reporting Register attribution sources Register attribution triggers Prioritize specific clicks, views, or conversions Define custom rules using filters Prevent duplication in reports Custom report windows Register multiple reporters Report schedules Constraints on Aggregation Reporting data Web-to-app and app-to-web measurement Understanding aggregation keys Understanding noise in summary reports Working with noise Contribution budget for summary reports
API updates
Private Aggregation API Private Aggregation API fundamentals Experiment and participate Unique reach measurement User demographic reporting K+ frequency measurement
Aggregation Service Summary reports
Private State Tokens
The Privacy Sandbox
What is the Privacy Sandbox? Proposal lifecycle in the Privacy Sandbox Timeline Glossary and key concepts Privacy Sandbox events Privacy Sandbox demos Blog Privacy Sandbox videos
Chrome-facilitated testing Relevance and measurement unified origin trial Enroll your site API status and feature releases
Privacy Sandbox testing FAQs Privacy-related compliance FAQs
Feedback
Prepare for phasing out third-party cookies Related Website Sets Related Website Sets: developer guide How Chrome evolved the First-Party Sets proposal Cookies Having Independent Partitioned State (CHIPS) Storage Partitioning Fenced frames
Federated Credential Management API Federated Credential Management API: developer guide Federated Credential Management API updates
Shared Storage Shared Storage API walkthrough
IP Protection Privacy Budget Bounce tracking mitigations
User-Agent reduction User-Agent reduction snippets
Topics API overview Topic classification Developer guide Integration guide Experiment and participate Topics API demos Test topic inference in a colab Topics API latest updates
Protected Audience API Experiment and participate
Protected Audience API: developer guide Seller guide: run ad auctions Buyer guide: join interest groups and generate bids Protected Audience API auction reporting Sequential auction setup with header bidding and multi-seller Protected Audience auction Improve Protected Audience API auction latency Opt-out of the Protected Audience API Report on Protected Audience API auction results Troubleshoot the Protected Audience API
Status of pending Protected Audience API capabilities Protected Audience API walkthrough
Content selection with Shared Storage Creative rotation Creative selection by frequency A/B Testing Known customers
Maximize ad relevance
Attribution Reporting overview Attribution Reporting end-to-end Experiment and participate Why Chrome plans to ship the Attribution Reporting API
Introduction to debug reports Set up debug reports Debugging cookbook
Attribution Reporting API developer guide Get started with Attribution Reporting Register attribution sources Register attribution triggers Prioritize specific clicks, views, or conversions Define custom rules using filters Prevent duplication in reports Custom report windows Register multiple reporters Report schedules Constraints on Aggregation Reporting data Web-to-app and app-to-web measurement Understanding aggregation keys Understanding noise in summary reports Working with noise Contribution budget for summary reports
API updates
Private Aggregation API Private Aggregation API fundamentals Experiment and participate Unique reach measurement User demographic reporting K+ frequency measurement
Aggregation Service Summary reports
Private State Tokens

Custom report windows

Custom report windows can provide faster event-level reports and selective reporting for aggregatable reports.

Published on • Updated on

Maud Nalpas
Maud Nalpas

Developer Relations Engineer for security/privacy, Chrome and web

Twitter GitHub

With custom report windows for Attribution Reporting (shipped in Chrome 110), you can instruct the browser to filter out conversions that occur after a certain time window.

  • For event-level reports, report windows allow you to get reports faster, in addition to filtering out conversions that take place after a certain time frame. Depending on your goal, another way to look at this is as a trade-off: using report windows gives you faster access to reports, but implies that you may lose some conversions. Review the event-level section for details.
  • For aggregatable reports, report windows allow you to filter conversions that take place after a certain time frame. Review the aggregatable section for details.

What is a custom report window?

A custom report window is the period after source registration during which reports of the corresponding type (event or aggregatable) may be created for this source.

For event-level reports, a report window also acts as the last window in which reports of the corresponding type will be sent.

Key points about reporting windows:

  • If no report_window is set, it falls back to expiry.
  • If no expiry is set, it falls back to the default of 30 days.
  • The minimum custom report window is one day. Even if you set a window of less than a day, the value will be clamped to a day. Review the examples for event-level reports.
  • A report window is defined in seconds.

Note that the way expiry works is unchanged; expiry changes which impressions or clicks are eligible for attribution.

Event-level reports: faster reporting and selective reporting

By default, event-level reports are sent following a schedule of built-in report windows: three windows for clicks and one window for views.

For event-level reports, custom report windows enable you to:

  • Get reports faster. Among other things, custom report windows enable you to get reports one day after the click or impression, instead of the two-day limit that was imposed by built-in expiry windows prior to Chrome 110.
  • Filter out conversions happening after a certain time frame.

To use report windows for event-level reports, add the event_report_window field to your Attribution-Reporting-Register-Source source registration header.

event_report_window

Event-level reports are scheduled to be sent at the latest shortly after the report window is reached.

Remember, in all the following examples, if the browser isn't running when a report is scheduled to be sent, the report is sent at browser startup. In this case, the browser may also delay some of these reports for a short random time on startup.

Schedule examples for event-level reports, clicks

expiry
event_report_window
If a conversion takes place one day after the click, browser sends report:Reports filtered out (not sent) for conversions taking place:
Default (30 days)(Not set)About 2 days plus 1 hour after click
Default reporting behavior — built-in window #1		About 30 days after click
Custom, set to 5 days(Not set)About 2 days plus 1 hour after click
Default reporting behavior — built-in window #1		More than 5 days after click
Default (30 days)Set to 1 dayAbout 1 day plus 1 hour after click

Custom behavior due to your custom report window. Note how this gets you this report faster than in the first row, where you need to wait for the first built-in window (2 days)		More than 1 day after click
Custom, set to 5 daysSet to 1 dayAbout 1 day plus 1 hour after click

Custom behavior due to your custom report window		More than 1 day after click
Default (30 days)Set to 4 hoursAbout 1 day plus 1 hour after click
Custom behavior due to your custom report window. This is due to clamping: the minimum report window is one day. Even if you set a report window to less than a day, for example, 4 hours, its value will be clamped to a day.		More than 1 day after click

Schedule examples for event-level reports, views

expiryevent_report_windowIf a conversion takes place one day after the click, the browser sends report:
Default (30 days)(Not set)About 30 days plus 1 hour after view
Default reporting behavior
Custom, set to 5 days(Not set)About 5 days plus 1 hour after view
Default reporting behavior
Default (30 days)Set to 1 dayAbout 1 day plus 1 hour after view
Custom behavior due to your custom report window
Custom, set to 5 daysSet to 1 dayAbout 1 day plus 1 hour after view
Custom behavior due to your custom report window

Aggregatable reports: selective reporting

For aggregatable reports, the benefit of report windows is only to filter out conversions happening after a certain time frame. There is no benefit in terms of getting reports faster.

To use report windows for aggregatable reports, add the following field to your Attribution-Reporting-Register-Source source registration header:

aggregatable_report_window

This will filter out conversions happening after a certain time frame.

Aggregatable reports are sent within an hour of trigger; they're randomly delayed between 10 minutes and an hour.

In the following table, for all cases, if a conversion takes place one day after the click or impression, the browser will send the report approximately an hour after the trigger.

expiry
aggregatable_report_window
Reports filtered out (not sent) for conversions taking place:
Default (30 days)(Not set)More than 30 days after click or view
Custom, set to 5 days(Not set)More than 5 days after click or view
Default (30 days)Set to 1 dayMore than 1 day after click or view
Custom, set to 5 daysSet to 1 dayMore than 1 day after click or view
Default (30 days)Set to 4 hoursMore than 1 day after click or view. This is due to clamping: the minimum report window is one day. Even if you set a report window to less than a day, for example, 4 hours, its value will be clamped to a day.

Engage and share feedback

You can participate and experiment with this API.

Updated on • Improve article

This site uses cookies to deliver and enhance the quality of its services and to analyze traffic. If you agree, cookies are also used to serve advertising and to personalize the content and advertisements that you see. Learn more about our use of cookies.