Google recently rolled out their highly anticipated Google Analytics 4 that leverages machine learning models to provide marketers with some unique tools that can help improve performance and drive results across Google Ads. Perhaps the most intriguing of these is the ability to create Predictive Audiences. Before diving too deeply into this article it must be noted that in order to take advantage of Predictive Audiences, you must upgrade to Google Analytics 4.
What is a predictive audience?
Located within the Audience Builder in Analytics, marketers are able to build audiences that leverage Predictive Metrics identified by Google’s machine learning to predict the future behavior of users in your target audience. There are currently three metrics that can be used to build Predictive Audiences:
- Purchase Probability: the likelihood that a user who was active in the last 28 days will make a purchase in the next 7 days.
- Churn Probability: the likelihood that a user who was active on your site will not be active in the next 7 days
- Revenue Prediction: the expected revenue within the next 28 days from a user who was active in the last 28 days
The accuracy of these metrics is directly related to the amount of data your Analytics account captures (volume of site traffic and website events). In order to be eligible for use, Google needs 1,000 positive and 1,000 negative outcomes (purchases or churned users) over a seven-day period. The sheer volume of data needed to create accurate audiences is potentially the largest barrier to entry for marketers who want to leverage Predictive Audiences.
Google currently provides suggestions for Predictive Audiences which include the following:
- Likely 7-day purchasers: Users who are likely to purchase in the next 7 days
- Likely first-time 7-day purchasers: Users who are likely to make their first purchase in the next 7 days
- Likely 7-day churning users: Active users who are likely to not visit your website in the next 7 days
- Likely 7-day churning purchasers: Purchasing users who are likely to not visit your site in the next 7 days
*Image credit: Google
How can I use a predictive audience?
Once a Predictive Audience has been built in Google Analytics 4, it is automatically added to the associated Google Ads account where it can be leveraged across the various campaign types the same way a customer list or custom audience can be used. Google outlines two obvious ways to use Predictive Audiences: in remarketing campaigns and in re-engagement campaigns. That said, at Parallel Path, we believe there are more than just two use cases for these audiences including strategic bid adjustments on Paid Search and prospecting campaigns.
By nature, remarketing campaigns are designed to serve ads to people who have already engaged with your website and are, therefore, more likely to convert than a cold audience. Predictive Audiences takes that idea to the next level by leveraging machine learning to identify more significant behavioral cues that form an audience of users who are even more likely to make a purchase. The ideal Predictive Audience to use in a remarketing campaign would be the “Likely 7-Day Purchasers” audience. A well-tailored message delivered at the right time might be exactly what this audience needs to complete a purchase.
Re-engagement campaigns are designed to keep people engaged with your business if they are showing signs of diminishing interest in your products, services, or content. By leveraging the “Likely 7-Day Churning Users” audience advertisers can reinforce their biggest value props through strategic messaging or special offers to keep users engaged with your business.
Strategic Bid Adjustments
Within Google Ads, advertisers are able to layer on audiences to their paid search campaigns and then place bid adjustments based on how they wish to prioritize these various audiences. By leveraging either the “Likely 7-Day Purchasers” or “Likely First-Time 7-Day Purchasers” audiences, search marketers can place higher bids on their keywords when they are triggered by someone who is more likely to complete a purchase than the standard user.
When attempting to grow your business, Prospecting Campaigns can be an excellent way to capture new customers. While this can be done by leveraging lookalike audiences or interest targeting, the “Likely First-Time 7-Day Purchasers” provides a more effective way of capturing new customers across GDN, Youtube, Gmail, and the Search Network.
In short, Google has now provided advertisers the ability to strategically reach the people who are the most likely to complete purchases, while also being able to keep people engaged who otherwise might never have come back to your site, both of which are extremely valuable when it comes to driving digital growth. It is extremely important to note that GA4 and these Predictive Audiences are very new and as such don’t have much data behind their effectiveness. That said, as more and more advertisers start leveraging Google’s machine learning tools, they will continue to get better and more accurate.