The Google Marketing Platform intends to use Chrome’s Privacy Sandbox to support key advertising use cases such as interest-based advertising, remarketing, and conversion measurement. For conversion measurement in particular, the platform will be using the Attribution Reporting API (ARA) in combination with advertiser-provided first-party data to measure conversions without third-party cookies.
Attribution Reporting API display ads experiment results
Google Marketing Platform teams have been conducting rigorous testing of the Privacy Sandbox APIs, in combination with other signals, to understand how the APIs can support ads performance when third-party cookies are not available.
In March 2024, we conducted an experiment on a portion of display ads traffic in Campaign Manager 360, Display & Video 360, and Search Ads 360 to assess conversion measurement reporting quality using the Attribution Reporting API. The following results are estimates at the specific time of the experiment and should not be taken as representative of the impact of varied third-party cookie availability on conversion measurement reporting quality. Since March 2024, there have been additional improvements made to the Privacy Sandbox ARA, technical improvements from Google Marketing Platform to better utilize the ARA, and outreach efforts to optimize adtech/publisher set up. The assumptions and results of the experiment do not account for Chrome’s announcement proposing an updated approach that elevates user choice rather than deprecating third-party cookies. We expect additional performance improvements compared to the results of this experiment due to the factors listed.
What we learned
Conversion reporting quality, as measured by Absolute Percentage Error (APE) to assess the accuracy of measuring conversions compared to third party cookies, varied by attribution type, such as view-through conversions (VTC) and click-through conversions (CTC). The experiment showed:
- VTC measurement, which is entirely dependent on the ARA, showed high APE, with the bulk of Campaign Manager 360 and Display & Video 360 campaigns falling in the 60% to 100% range when analyzing traffic both already integrated with the ARA and all traffic. However, when analyzing only the subset of traffic already integrated with the ARA, VTC and CTC results show improvements. VTC results for this integrated traffic segment showed that 65% of campaigns fell within the 60-100% APE range, compared to 85% of campaigns when considering all traffic.1
- CTC measurement, which are only partially reliant on the ARA due to the presence of first-party cookies, showed significantly better APE, particularly on Search Ads 360, where the vast majority of campaigns fell in the 0%-20% APE.1
While the results should not be taken as fully representative of the true impact of varied third-party cookie availability in the future, there are several factors that contributed to conversion measurement reporting quality in this experiment. First, it is important to note that the Google Marketing Platform depends on third party ad tech providers to support the use of the attributionsrc parameter in impression and click trackers so marketers can receive conversion attribution reporting downstream. That’s why we are working with the industry to encourage adoption. Additionally, the ARA has limits to how many conversion events can be reported after a click or impression takes place. Teams expect that impact will be further mitigated as customers optimize their conversion setup to account for the limitations in place. Additionally, these results do not account for Chrome’s proposal for an updated approach to third-party cookies. We expect that the conversion measurement quality will improve as we continue these efforts and as changes are incorporated.
Looking ahead
Google Marketing Platform teams will continue iterating on our ARA integration in consultation with the Competition and Markets Authority in the UK, and share regular feedback on the API as it evolves. With the rapid advancements in AI, first-party data solutions and privacy-preserving innovation, publishers, ad tech providers, search engine partners, and marketers have the tools to succeed.
1. Google Internal Data. Results are based on the standalone Floodlight ARA experiment conducted in March 2024.
How the Attribution Reporting API works
The Attribution Reporting API enables measuring two events that are linked together: an event on a publisher's website, such as a user viewing or clicking an ad, with a subsequent conversion on an advertiser site.
Event-level reports
Event-level reports are generated when the browser matches clicks or views with conversion data defined by an ad tech. Later, the browser sends the resulting reports to a predefined endpoint, with some delay and noise.
Summary reports
Summary reports are generated as follows:
- A user clicks or views a specially configured ad. The browser on the user's local device records this event, alongside pre-specified attribution configuration data.
- When the user converts, the browser matches this detailed click or view event (known as the attribution source event) with detailed conversion data (known as attribution trigger data). The dimensions of detail captured are pre-defined by an ad tech company, and the browser follows specific logic that is defined by the ad tech. The browser outputs this data in an aggregatable report.
- Aggregatable reports are encrypted by the browser and sent to an ad tech server. From the ad tech server, the aggregatable reports are sent to the aggregation service to produce a summary report.
- Summary reports are then made available to the ad tech. Note that summary reports are not delayed to the same extent as event-level reports.
Learn more about Attribution Reporting for Web overview.
Next steps for publishers, AdTech, and search engines
Google Marketing Platform ecosystem partners should take the following steps to verify ad event registration with the Attribution Reporting API, enabling marketers to gain valuable conversion measurement insights:
Include the attributionsrc parameter in impression and click trackers
Publishers, ad techs, and search engines should ensure that the Attribution Reporting API’s attributionsrc HTML parameter is maintained within impression and click trackers. This will support the use of Floodlight conversion tags for measurement and bidding across the Google Marketing Platform. Current testing of the Attribution Reporting API shows the attributionsrc parameter is frequently missing before Google Marketing Platform impression or click tags fires. Ecosystem partners should confirm that provided attributionsrc attributes are included in ad tags so marketers can receive attribution for these ad events via the Attribution Reporting API.
Add the attributionsrc parameter to impression & click events in VAST
For publishers and SSPs that are integrated with Google’s Interactive Media Ads SDK, support for the attribute is automatically included. For publishers and SSPs that are integrated with other SDKs, work with your SDK to adopt the VAST addendum published by the Interactive Advertising Bureau (IAB) that enables support for the Attribution Reporting API. Regardless of how the VAST protocol is being used, it is important to note that registration must occur in order for impression and click events to be eligible for attribution.
Call ads and conversions from HTTPS
Ads and conversion tags should be rendered on or called from HTTPS pages or iframes. If these tags are called via a URL redirect, then the URL sending the redirect should also be using HTTPS.
Next steps for Marketers
Use the Google tag as the format for your Floodlight activities
Google Tag or Google Tag Manager allows you to fortify click-through conversion attribution measurement capabilities in an environment with fewer deterministic signals. This tagging infrastructure makes it possible to gather a click ID for deterministic linking of a click event with a conversion event. If a Google Tag solution is already deployed for other Ads products, you can add the Campaign Manager 360 or Display & Video 360 snippet to your existing tags to measure ad events for the Google Marketing Platform.
Disable attribution for Floodlight tags where they aren't needed for bidding or optimization, or change the counting method from Standard to Unique for activities driving many repeat conversions per user
The Attribution Reporting API has limits to how many conversion events can be reported after a click or impression takes place. This privacy constraint imposed by the API, otherwise known as the Conversion Contribution Limit, specifically limits the advertiser to measure 20 conversion events after an ad interaction. In order to make the most of the limit, it is recommended that you deactivate any Floodlight tags that are not essential for your measurement needs. Additionally, consider converting Standard tags that experience high volumes of traffic on specific pages of your website to Unique tag types. This approach will allow you to preserve full funnel conversion attribution measurement capabilities. Please note that this limitation is the number of conversions per user and not the number of conversion tags that are triggered. You should choicefully select which conversion tags are enabled to preserve conversion measurement throughout your site funnel.
Create a strong framework to collect and maintain users’ consent with Consent Mode
If a user from the European Economic Area is using your website or app and you measure user behavior with Google tags, you need to collect and pass end-user consent choices to Google. Consent mode allows you to adjust how your Google tags behave based on the visitor's interaction with the consent banner on your website.
Confirm your conversion domains are set up properly for each placement in Campaign Manager 360
The Attribution Reporting API supports conversion attribution on three domains, or otherwise known as conversion destinations, after an impression or click event. While Campaign Manager 360 will infer the top three domains on your website, you can update those domains at the placement level to instruct the API where to report on conversions when an ad event takes place. Conversion attribution will not be possible using the Attribution Reporting API if the appropriate conversion domains are not specified at the time of the ad event.