Target Google Ads audiences based on new predicted purchase and churn metrics in Google Analytics
Available in App + Web properties, automate audience creation based on predicted behavior.
Applications of machine learning have brought us into the era of predictive marketing in which bids, ads and targeting are optimized based on predicted outcomes. The latest sign of this in Google Analytics is the addition of predictive metrics and audiences in App + Web properties.
When building audiences in an App + Web property in Google Analytics, you’ll see suggestions for predictive audiences using either of these metrics:
- Purchase Probability: Users deemed most likely to make a purchase within the next seven days. The idea here is to be able to identify and target more than just abandoned cart users and reach those visitors that didn’t put an item in their carts but are seen as likely to make a future purchase.
- Churn Probability: Active users deemed unlikely to visit your site or app in the next seven days. These are past site visitors that aren’t likely to come back to your site on their own within the week.
There are two predictive audience options based on these new metrics:
- Purchase Probability audiences: Likely 7-day purchasers and Likely first-time 7-day purchasers.
- Churn Probability audiences: Likely 7-day churning purchasers and Likely 7-day churning users. Give past site visitors reason to come back.
Analyze using predictive metrics. Using the Analysis module, you can use these new predictive metrics to build reports such as understanding which marketing campaigns brought in users deemed to have the highest Purchase Probability.
Why we care. Machine learning-driven audiences aren’t new in Google Analytics — Smart Lists debuted way back in 2014 — but these updates offer more options and flexibility. Expect more to come. Google says the new metrics and audiences will be available in the coming weeks. You’ll need to have purchase events set up or automatically measuring in-app purchases in your App + Web property.
Opinions expressed in this article are those of the guest author and not necessarily Search Engine Land. Staff authors are listed here.