The Measurement Simulation Library
The Measurement Simulation Library helps you to understand the impact of your
Privacy Sandbox integration by presenting historical data as if it were
collected by the Attribution Reporting API. This lets you to compare your
historical conversion numbers with Measurement Simulation Library results to see
how reporting accuracy might change. You can also use the Measurement Simulation
Library to experiment with different aggregation key structures and batching
strategies, and train your optimization models on Measurement Simulation Library
reports to compare projected performance with models based on current data.
Read the Measurement Simulation Library design proposal to learn more.
Except as otherwise noted, the content of this page is licensed under the Creative Commons Attribution 4.0 License, and code samples are licensed under the Apache 2.0 License. For details, see the Google Developers Site Policies. Java is a registered trademark of Oracle and/or its affiliates.
Last updated 2024-11-14 UTC.
[null,null,["Last updated 2024-11-14 UTC."],[[["The Measurement Simulation Library uses historical data to demonstrate how the Privacy Sandbox's Attribution Reporting API might impact your conversion tracking."],["By comparing current data with simulated reports, you can assess potential changes in reporting accuracy."],["The library enables experimentation with different aggregation keys and batching strategies to optimize campaign performance."],["You can train optimization models using simulated data for comparing projected performance against existing models based on current data."]]],["The Measurement Simulation Library allows users to analyze the impact of Privacy Sandbox integration. It simulates Attribution Reporting API data using historical information, enabling comparison of past conversions with simulated results. This allows for assessment of reporting accuracy changes. Users can also test varied aggregation key structures and batching strategies. Finally, the library can be used to train optimization models, allowing performance comparison between models using simulated data and those relying on existing data.\n"]]
