What are predictive metrics in Google Analytics 4 (GA4)

Table of Contents

  1. Predictive metrics overview
  2. Prerequisites for predictive metrics
  3. Using predictive metrics
  4. Summary

In this article, I am going to talk about predictive metrics in Google Analytics 4 (GA4).

Predictive metrics overview

Google Analytics 4 supports predictive metrics, which are derived through machine-learning algorithms that measure the progress to a conversion. GA4 automatically enriches your data by using Google machine-learning algorithms on your dataset to predict the future behavior of the users.

With predictive metrics, you can identify users and their actions on the website that will likely lead to a purchase or conversion. This will help you to discover more users who can purchase the product in the next 7 days.

There are currently three predictive metrics supported by GA4.

  1. Purchase probability:
    The probability that a user who was active in the last 28 days will generate a purchase event within the next 7 days. Currently, only purchase, ecommerce_purchase and in_app_purchase events are supported in GA4.
  2. Churn probability:
    The probability that a user who was active on your website within the last 7 days will not be active within the next 7 days.
  3. Revenue prediction:
    The revenue expected from all purchase events within the next 28 days from the users who were active in the last 28 days on the website.

Get the E-Book (50 Pages)

Get the FREE E-Book (50+ Pages)

Prerequisites for predictive metrics

Since predictive metrics are derived from Google machine-learning algorithms, there are a few prerequisites to successfully training predictive models, as follows.

  • To get eligible for predictive metrics like ‘purchase probability’ and ‘churn probability’, you must configure the purchase event and send it to the GA4 property.
  • A minimum number of 1000 positive and negative samples (purchasers and churned users) are required. This means there should be at least 1000 users who have triggered the predictive condition to purchase and 1000 users who did not.
  • Model quality (regular traffic generating purchase events) must be sustained over a period of time, which is generally 28 days.

Predictive metrics for each eligible model will be generated for each active user once per day. In the case that any of the prerequisites are not met, or fall below the minimum threshold of users, then Google Analytics 4 will stop updating the predictive metrics and it will be unavailable in Google Analytics.

You can check the status of each prediction provided by GA4 in the predictive section within ‘Suggested audiences’ templates in the ‘Audience builder’.

If there is not sufficient data to use predictive modeling, an audience will be marked as ‘Not eligible to use’.

Using predictive metrics

You can use predictive metrics in ‘Audience builder’ while creating a custom audience.

Using predictive metrics, you can create the following type of custom Audiences

Likely 7- day purchasers

Users who are likely to purchase in the next 7 days. The predictive metric used in this audience is:

Purchase probability greater than the 90th percentile (more than 90%).

Likely first-time 7-day purchasers

Users who are likely to make their first purchase in the next 7 days. The predictive metric used in this audience is:

Purchase probability greater than the 90th percentile (more than 90%)

AND

LTV (user lifetime value) equals zero.

Likely 7-day churning users

Active users who are likely to not visit your website in the next 7 days. The predictive metric used in this audience is:

Churn probability is greater than 80th percentile (more than 80%).

 

Likely 7-day churning purchasers

Purchasing users who are likely to not visit your website in the next 7 days. The predictive metric used in this audience is:

Churn probability greater than 80th percentile (more than 80%)

AND

LTV (user lifetime value) greater than zero

OR

Purchase event (users who have done purchase)

OR

Ecommerce_purchase (purchase event for mobile application).

 

You can also use predictive metrics in the ‘Analysis’ tab while creating the User Lifetime report. Follow the below steps to create a User Lifetime report using predictive metrics.

Step 1: Navigate to your Google Analytics 4 property and click on the ‘Analysis’ drop-down menu from the reporting menu.

Step 2: Click on ‘Analysis Hub’.

Step 3: You will see the various analysis templates. Click on the ‘+’ sign to select the blank template.

Step 4: A new console will open. Now click on the drop-down available in ‘Tab settings’ under ‘Technique’.

Step 5: Select ‘User lifetime’.

Step 6: By default, predictive metrics are not added to the report and you need to add them manually.

Click on the ‘+’ sign under ‘Metrics’ in the ‘Variables’ tab.

Step 7: An overlay will appear on the right-hand side like below. Click on ‘Predictive metrics’ from the list available.

Step 8: The ‘Predictive metrics’ drop-down will expand, like below.

All predictive metrics are divided further into sub metrics. If you click on any of the metrics it will select all the sub metrics as well.

To give you more idea about the sub metrics, they show the probability at which users are assigned to the main predictive metric.

Churn probability:

  • 10th percentile: 10% probability that a user who was active on your website within the last 7 days will not be active within the next 7 days.
  • 50th percentile: 50% probability that a user who was active on your website within the last 7 days will not be active within the next 7 days.
  • 80th percentile: 80% probability that a user who was active on your website within the last 7 days will not be active within the next 7 days.
  • 90th percentile: 90% probability that a user who was active on your website within the last 7 days will not be active within the next 7 days.
  • Average: Average probability that a user who was active on your website within the last 7 days will not be active within the next 7 days.

Purchase probability:

  • 10th percentile: 10% probability that a user who was active on your website within the last 28 days will purchase within the next 7 days.
  • 50th percentile: 50% probability that a user who was active on your website within the last 28 days will purchase within the next 7 days.
  • 80th percentile: 80% probability that a user who was active on your website within the last 28 days will purchase within the next 7 days.
  • 90th percentile: 90% probability that a user who was active on your website within the last 28 days will purchase within the next 7 days.
  • Average: Average probability that a user who was active on your website within the last 28 days will purchase within the next 7 days.

The same goes for in-app purchase probability, but this is only related to mobile apps, and not websites.

Now select checkboxes in front of  ‘Churn probability’ and ‘Purchase probability’ and then click on ‘Apply’.

Step 9: You see the selected predictive metrics are available in the ‘Metrics’ list under the ‘Variable’ tab.

Now, double-click on any of the predictive metrics to add it to the report. For example, here I am adding ‘Purchase probability: Average’ and ‘Churn probability: Average’.

Congratulations! You have successfully created a User Lifetime report using predictive metrics. Your report will look similar to the one below.

Summary

Imagine you are running an online website and want to drive more sales through the digital medium. Predictive analytics will help you understand the user base which is most likely to purchase your products. You can also add the predictive audience list into Google Ads to engage with them more.

Another example could be that you are a blogger and plan to publish some new blog posts. In this case, predictive metrics such as ‘churn probability’ will give you the details of users who are not likely to visit your blog in the coming 7 days. If the numbers are high, you can consider publishing the blog posts a week later or run a Google Ads campaign to encourage them to read your blog.

Register for the FREE TRAINING...

"How to use Digital Analytics to generate floods of new Sales and Customers without spending years figuring everything out on your own."



Here’s what we’re going to cover in this training…

#1 Why digital analytics is the key to online business success.

​#2 The number 1 reason why most marketers are not able to scale their advertising and maximize sales.

#3 Why Google and Facebook ads don’t work for most businesses & how to make them work.

#4 ​Why you won’t get any competitive advantage in the marketplace just by knowing Google Analytics.

#5 The number 1 reason why conversion optimization is not working for your business.

#6 How to advertise on any marketing platform for FREE with an unlimited budget.

​#7 How to learn and master digital analytics and conversion optimization in record time.



   

My best selling books on Digital Analytics and Conversion Optimization

Maths and Stats for Web Analytics and Conversion Optimization
This expert guide will teach you how to leverage the knowledge of maths and statistics in order to accurately interpret data and take actions, which can quickly improve the bottom-line of your online business.

Master the Essentials of Email Marketing Analytics
This book focuses solely on the ‘analytics’ that power your email marketing optimization program and will help you dramatically reduce your cost per acquisition and increase marketing ROI by tracking the performance of the various KPIs and metrics used for email marketing.

Attribution Modelling in Google Analytics and Beyond
Attribution modelling is the process of determining the most effective marketing channels for investment. This book has been written to help you implement attribution modelling. It will teach you how to leverage the knowledge of attribution modelling in order to allocate marketing budget and understand buying behaviour.

Attribution Modelling in Google Ads and Facebook
This book has been written to help you implement attribution modelling in Google Ads (Google AdWords) and Facebook. It will teach you, how to leverage the knowledge of attribution modelling in order to understand the customer purchasing journey and determine the most effective marketing channels for investment.

About the Author

Himanshu Sharma

  • Founder, OptimizeSmart.com
  • Over 15 years of experience in digital analytics and marketing
  • Author of four best-selling books on digital analytics and conversion optimization
  • Nominated for Digital Analytics Association Awards for Excellence
  • Runs one of the most popular blogs in the world on digital analytics
  • Consultant to countless small and big businesses over the decade
error: Alert: Content is protected !!