Select Language:
A month ago, Digital Phablet highlighted a new help document from Google Ads detailing the Performance Max optimization experiments: A/B testing assets beta. Recently, some advertisers have started to see these features live, allowing for controlled A/B tests within campaigns.
Matteo Braghetta shared this on LinkedIn (credit to Adrian Dekker for the discovery) and explained that Google is now enabling controlled A/B testing for creative assets directly inside Performance Max. Advertisers can test two different asset sets within a single asset group, dividing traffic according to a specific split and organizing the experiment for better insights.
In the Google Ads interface, this feature appears as an option to set up these experimental splits. The process involves selecting a Performance Max campaign and choosing one asset group, then defining the experimental parameters:
– Assets A serve as the control, representing the current live assets.
– Assets B are the treatment, comprising new or alternative creatives.
– Shared assets are common across both variants, ensuring consistent elements.
– The traffic split, such as 50/50, determines how much traffic goes to each asset set.
– Once the setup is complete, the experiment is launched and allowed to run for the recommended period, typically around 4 to 6 weeks.
– Afterward, results are analyzed, and the most effective asset set can be applied permanently to the asset group.
This new testing capability aims to streamline creative optimization within Performance Max campaigns, providing more data-driven insights for campaign performance improvements.
ChatGPT
Perplexity
Gemini AI
Grok AI
