Getting started with Google’s Privacy Sandbox: A comprehensive overview

⚠️ Disclaimer: This blog post was last updated in April 2023. Please note that it may contain some information that has to be revised in accordance with Google’s latest announcements.

Privacy is a key factor in today’s mobile marketing world, driving changes and challenges critical to marketers and developers alike. At the forefront of this trend are Apple and Google, two of the leading ad platforms in the market. They are most notably devising ways to measure ad performance without tracking users. In 2018, Apple introduced its approach to privacy-preserving attribution, called SKAdNetwork, and has continued to update it ever since.

Google also began a beta rollout of the Privacy Sandbox earlier this year, signaling yet another huge change from the contemporary marketing measurement methods. Expected to be fully revealed by 2024, this initiative aims to create a new set of advertising standards that will enable mobile growth while protecting user privacy.

The Privacy Sandbox is expected to completely shift digital marketing. What will privacy-safe marketing look like? Will retargeting die without device identifiers? How will mobile apps have to measure ad effectiveness? Answering these questions starts with understanding Privacy Sandbox’s fundamentals.

What does the Privacy Sandbox mean for marketers?

The gist of the Privacy Sandbox is allowing various marketing activities while deprecating the major user identifiers. While many have long been forecasting the demise of third-party cookies on Chrome, it was a different story for GAID.

Similar to Apple’s Identifier for Advertisers (IDFA), GAID is a unique device identifier that allows mobile apps to anonymously track user activity in the Android ecosystem. In order to measure ad performance, mobile measurement partners (MMPs) like Airbridge rely on these identifiers and track individual users’ in-app activities such as installs and purchases.

However, GAID is fading into history along with the advent of Google’s Privacy Sandbox. This is in fact one of the biggest differences in the privacy policies of Google and Apple. Apple, according to its App Tracking Transparency framework, does allow advertisers to access the IDFAs of users who have opted into app tracking. Google, on the other hand, will no longer provide GAID-based deterministic attribution regardless of user consent. Considering that Android is the leading mobile operating system with a 71.8% global market share, the initiative is bound to have a huge impact on mobile marketing.

Google has not announced when exactly it will phase out GAID yet, but the company rolled out the first Beta for the Privacy Sandbox on Android in February 2023. In addition, Google is revealing more and more details about the Privacy Sandbox with time.

How can the Privacy Sandbox and SKAdNetwork be compared?

Google’s Privacy Sandbox is similar to Apple’s SKAdNetwork in that it allows mobile apps to measure the success of their marketing campaigns while maximizing user privacy. However, their ways of polishing and delivering the end result are vastly different.

In an effort to best preserve each user’s privacy, SKAdNetwork data is delayed, with the earliest SKAdNetwork postback sent approximately 24 hours after install. Similarly, the Privacy Sandbox also ensures individual users cannot be identified. This pushes Android marketers to define their campaign objectives up-front, before the ad is served, and that only such predefined datasets will become available mainly in aggregated forms.

Still, there is a bright side: these aggregated reports with richer, high-fidelity data are sent with a random delay of only one hour or even shorter. This is clearly in contrast to how SKAdNetwork provides data in a much more time-restricted manner. Moreover, to trigger the SKAdNetwork postback, the privacy threshold should be met, meaning that a marketing campaign needs to hit a certain number of daily app installs. The Privacy Sandbox does not impose such restrictions.

Instead, to minimize the risk of identifying a user, the Privacy Sandbox uses noise. A random amount of data that does not substantially affect its analysis results would be added to the aggregated reports. In other words, noise supports private measurement without preventing marketers and developers from making informed decisions.

Does this signal the end of an era?

Fortunately, it is still possible to measure performance using the tools we are used to. For instance, the functionality provided by the Play Install Referrer will not be impacted by the Privacy Sandbox. A string of numbers will continue to be sent to the Google Play Store whenever an ad click occurs to allow attribution providers to measure the performance of install campaigns. Some other types of identifiers are also used by MMPs, which will employ all available means to keep delivering advanced measurement solutions.

Furthermore, the Privacy Sandbox proposes a number of features to support privacy-centric mobile marketing. As it will become increasingly difficult to collect data and identify target users to serve ads based on real-time trends, Google will  provide a set of APIs, each designed to support different marketing activities. Some of the main ones and their objectives are as follows:

.tg  {border-collapse:collapse;border-spacing:0;} .tg td{border-color:black;border-style:solid;border-width:1px;font-family:Arial, sans-serif;font-size:14px;  overflow:hidden;padding:10px 5px;word-break:normal;} .tg th{border-color:black;border-style:solid;border-width:1px;font-family:Arial, sans-serif;font-size:14px;  font-weight:normal;overflow:hidden;padding:10px 5px;word-break:normal;} .tg .tg-u5bg{border-color:inherit;font-family:Tahoma, Geneva, sans-serif !important;text-align:left;vertical-align:top} .tg .tg-og15{background-color:#EFEFEF;border-color:inherit;font-family:Tahoma, Geneva, sans-serif !important;font-weight:bold;  text-align:left;vertical-align:top}                                                

Privacy Sandbox Proposals

Overview

Topics API

Enable interest-based advertising without relying on user-level data

Protected Audience API

Show customized ads to users based on their app interactions without third-party data sharing; previously known as FLEDGE

Attribution Reporting API

Measure ad performance while limiting user-level information sharing through event-level reports

How Airbridge is already in action

Without a doubt, Google’s Privacy Sandbox is a game-changer; it will be reshaping the mobile marketing landscape. Saying goodbye to GAID indicates the need to reimagine performance measurement. To stay ahead of the game, it is crucial to ponder over how to develop your marketing strategy using the latest tips and tricks from Google.

Airbridge has already set out to work on the Privacy Sandbox. As an App Attribution Partner (AAP) of Google, we are strategizing from multiple angles to improve our current attribution model. We also offer other measurement methods such as marketing mix modeling (MMM), presenting a more holistic view.

There are endless twists and turns, but you don’t have to pass through them by yourself. Airbridge will be at your side to navigate the future of measurement.