Segment with Einstein Engagement Scoring

Learning Objectives

After completing this unit, you’ll be able to:

  • Activate Einstein Engagement Scoring.
  • Identify engagement personas.
  • Establish persona-based campaigns.
  • Use Einstein Splits in Journey Builder.

Before You Start

Before you start this module, make sure you complete Einstein for Marketing Basics. The work you do here builds on the concepts and work you do in that badge. Also, talk to your account manager about requirements for using Einstein features within Marketing Cloud. 

Einstein in Marketing Cloud

Artificial intelligence (AI) is changing the landscape of email marketing, and we want to give you the latest and greatest Marketing Cloud tools to bring the intelligence of Salesforce Einstein to your marketing. As a refresher, Salesforce Einstein is Salesforce’s AI brand, and includes all of our intelligent products or services. In this module, Einstein—the brilliant scientist—becomes a brilliant email marketer. With his help, you learn how to improve segmentation, create personalized content and recommendations, and optimize how and when you engage with your customers. 

Einstein Engagement Scoring

Our AI journey begins with Einstein Engagement Scoring, a great way to get to know your audience. Einstein Engagement Scoring is a data science–driven feature that helps you target customers to increase engagement and conversions for your email program. This scoring model uses customer data and machine-learning to assign subscribers a score that predicts their likelihood to open an email, click a link, stay subscribed, or convert once they’ve landed on your website. You can use these scores to do some pretty cool things.

  • Predict engagement. Track, segment, and target audiences for specific campaigns.
  • Gain audience insights. Understand the factors that predict engagement, spot trends, and measure audience health.
  • Take action on the data. Use scores to create audiences, target persona segments, split paths in Journey Builder, and personalize content.

Activate Einstein Engagement Scoring

In this module, we assume you are a Marketing Cloud administrator with the proper permissions to activate Einstein features. If you’re not an administrator for Marketing Cloud, that’s OK. Read along to learn how your administrator would take the steps in a production org. Don’t try to follow these steps in your Trailhead Playground. Marketing Cloud isn’t available in the Trailhead Playground. 

So let’s begin. 

  1. Navigate to Einstein.Hovering over Einstein navigation and selecting Einstein Engagement Scoring.
  2. From the dropdown, select Einstein Engagement Scoring.
  3. Click Activate Einstein Engagement Scoring. Screenshot of an inactivated account with Activate Einstein Engagement Scoring circled.

If you already have 90 days of email engagement data in your account, then you are ready to go. If not, you have to wait until you do. Start that countdown clock. 

Once your account is ready, you receive an email notification (around 48 hours after you’ve activated and have the required data). Then, a new data extension is created in a JourneyIQ folder called MC_Einstein_Predictive_Scores. You can use this data extension to build filtered data extensions for segmentation and specific persona-based campaigns. Stay tuned, we cover personas later in this module. 

Einstein Engagement Scoring Dashboard

After activation, you can head to the Einstein Engagement Scoring dashboard under Einstein to check out a variety of predictions. Let’s review each tile of the dashboard. 

ngagement Scoring dashboard with callouts for 5 sections.

(1) Email Engagement Prediction: Displays the size of your population, along with the number of subscribers that fall into one of these personas: win-back/dormant, window shoppers, selective subscribers, or loyalists. We define these personas in the next section. 

(2) Subscriber Retention Prediction: Displays the likelihood of your subscribers to stay subscribed to your emails in the next 14 days.

(3) Email Open Prediction: Displays the likelihood your subscribers open an email in the next 14 days, along with audience health. 

(4) Email Click Prediction: Displays the likelihood your subscribers click an email in the next 14 days, along with audience health. 

(5) Web Conversion Prediction: Displays the likelihood your subscribers convert (make a purchase or download content) in the next 14 days.

Here is a close up of the Email Open Prediction tile. 

Email open predictions with model confidence and percent change since last week circled.

Each tile shows a percent change from the previous week, helping you determine if the scores are improving or if you need to make changes. You may also wonder: Are blue dots a good thing? Yep. Model confidence or model accuracy is shown for each prediction—and the more dots, the better. Note that fair or poor model confidence may indicate that there is not enough subscriber data for predictions.



Click View Details on any tile in the dashboard to get more detail about the predictions.

Engagement Personas

Einstein Engagement Scoring uses personas to group subscribers based on predicted opens and clicks. These personas are updated daily. Let’s review how you can use these personas for segmentation and targeted campaigns.

Desired Customer Behavior
How to Target
Subscribers with high open and click engagement.
  • Refer-a-friend
  • Increase engagement
  • Purchase more
  • Create a lookalike audience based on your best-performing subscribers.
  • Increase sending frequency.
  • Send exclusive or limited offers.
  • Promote an event to encourage signups.
Window Shoppers
Subscribers with high open and low click engagement.
  • Click a call to action (CTA)
  • Send personalization and recommendations.
  • Test different CTAs.
  • Try targeted incentives.
Selective Subscribers
Subscribers with low open and high click engagement.
  • Open an email
  • Test subject line optimization.
  • Personalize based on past purchases.
Subscribers with low open and click engagement.
  • Avoid unsubscribe
  • Reengage

  • Send a preference center survey.
  • Try a different channel, like Advertising Audiences or SMS.
  • Run a win-back campaign with an offer.

Einstein Splits

Even better, you can use these personas and scoring predictions in Journey Builder to maximize subscriber engagement and conversions throughout a journey. How? By using Einstein Splits. Let’s review the five Einstein Split options in Journey Builder.

When to Use
Persona Split
Engage customers based on their Einstein Engagement Scoring Personas (loyalist, window shoppers, selective subscribers, and win-back/dormant).
Conversion Split
Engage customers based on their likelihood to make a purchase, download content, or complete a form on your website. (Note: You need to add collect tags to your website before you can use conversions.)
Email Open Split
Engage customers based on their likelihood to open an email.
Email Click Split
Engage customers based on their likelihood to click a link.
Retention Split
Engage customers based on their likelihood to continue as a subscriber.

Let’s walk through an example of how Einstein Splits could be used by shoe retailer, Cloud Kicks. Cloud Kicks recently activated Engagement Scoring, and they are interested in using those scores in Journey Builder. Yasmin Ahmad, its digital marketing manager, wants to use a Persona Split in an upcoming winter sale campaign to target high-value customers. For those subscribers less likely to engage, she wants to entice them with a coupon. Let’s follow along as she creates this journey and adds the Persona Split. 

Yasmin navigates to Journey Builder and clicks Create New Journey and then selects Create Journey From Scratch.  

New journey setup screen.She then drags over the Data Extension entry source and selects the Cloud Kicks customer data extension. Next, she drags over the Einstein Split icon to the journey window.

Journey dashboard with Einstein Split.

Yasmin then double clicks the Einstein Split icon and selects the type of Einstein Split she plans to use, Persona Split, and then clicks Done.

elect an Einstein Split Type from the list of options.

Yasmin decides to send the coupon email to loyalists and selective subscribers if they didn’t click the original email after 3 days. Window shoppers and win-back/dormant subscribers receive a reminder SMS message after 3 days, and the remainder of the audience gets targeted by a social media ad campaign. Here is an example of how her finished campaign looks in Journey Builder. 

Journey dashboard with Einstein Split.

With these Einstein tools, you can create an engaging multi-channel campaign that uses tailor-made strategies based on a subscriber’s persona. In the next unit, we create personalized content using Einstein features.  

Rights of ALBERT EINSTEIN are used with permission of The Hebrew University of Jerusalem. Represented exclusively by Greenlight.