Everything you wanted to know about the Einstein Engagement Scoring Feature

This is the second article in the series of articles exploring the Salesforce Marketing Cloud Einstein feature. We have started the series with the first article featuring the Einstein Content Selection feature (divided into part 1 and part 2).

In this blog, we shall be focusing on Einstein Engagement Scoring from the house of Einstein- The Artificial Intelligence [AI] product from Salesforce Marketing Cloud.  

Einstein Engagement Scoring predicts the probability of the contact or subscriber [for email] to engage with your marketing content like Emails & Mobile push messages. In the background, Einstein uses the existing customer data stored in Marketing Cloud and its machine learning capabilities to generate predictive models. These models then assign scores to all the contacts within Salesforce Marketing Cloud. These scores predict the likelihood of the contact to open or click the emails and engage with push notifications. 

Einstein Engagement Scoring comes as a part of the following Editions in Salesforce Marketing Cloud: 

• Corporate Edition  
• Enterprise Edition  

Additionally, you can purchase this as an add-on to the Professional Edition of Marketing Cloud. This feature needs to be activated from the SetUp option in Marketing Cloud. 

Benefits of Einstein Engagement Scoring 

Einstein Engagement Scoring provides multiple benefits. Let’s explore a few: 

• It helps in discovering the next action of the contact. It adds scores for every contact on the likelihood of open, clicks an email. This means fewer chances of unsubscribes and more chances of conversion. 
  
• Another advantage is that it helps to understand the factors that drive engagement of the contacts. It also helps in identifying trends in the industry and measuring audience health. You get a perspective on the subscriber engagement with the brand. 
  
• Last but not the least, you get segments within the Marketing Cloud which can help to target various personas. This segmentation can be used in Journey Builder with the help of Einstein Split Activity. This aids in choosing the right contact for the right path based on various personas, eventually generating better engagement with the customer and resulting in an enhanced customer experience.  

How it works  

The Einstein Engagement Scoring feature predicts how a particular contact will engage with your brand and their likelihood of making a purchase. Now, this analysis is done based on the contact’s email engagement data for the past 90 days. This data includes the number of emails sent to that subscriber, clicks, opens, etc..  

Einstein also considers other factors to determine the subscriber engagement with the emails. These factors can range from email clients used by the contact to the device used to access emails to the frequency of emails opens. These factors clubbed with the engagement data of the subscriber help Einstein to predict the likelihood of the subscriber opening the email.  

Einstein then groups these subscribers into various personas based on the number crunching as detailed above: 

• Loyalists: These are your best bet kind of subscribers and they have a high probability of clicking and opening your emails. 

• Window Shoppers: As the name suggests, these subscribers would open the emails but will have low click engagement. 

• Selective Subscribers: These are subscribers who are choosy and generally have a low open rate; the click engagement, however, is high. 

• Winback/Dormant: These are the subscribers who show a low open rate as well as low click engagement; not good news at all. 

These personas are used to make prediction models over the next 14 days for the subscribers who are most likely to open and click the emails and hence more likely to make a purchase and stay subscribed. 

Salesforce Marketing Cloud recalculates the subscriber scores on a Business Unit level to make sure that the predictions are made for BUs which are actively sending the emails. We don’t want predictions made on BUs which are used as sandboxes or used as Admin functions. 

Einstein Engagement Scoring feature saves the engagement data of subscribers in the below data extension: 

• Einstein_MC_Predictive_Scores  
• Einstein_MC_MobilePush_Scores 

These data extensions can be accessed via Email Studio or Contact Builder within Salesforce Marketing Cloud. 

Einstein Engagement scoring Use cases 

Based on the user personas, there could be various use cases that will fit the scenarios and are listed below: 

• Loyalists: These are your most prized customers and you should target them with special deals or offers to strengthen your relationship with them.  

• Window Shoppers: These shoppers are ready to open your emails but still not ready to take the next step. In this scenario, it is recommended to target them with a specific campaign as per their liking along with specific keywords in the subject and body of the email to grab their attention and get them to click on the CTA. 

• Selective Subscribers: These subscribers are choosy and hence you should target them with personalized offers which will nudge them to open the email. 

• Winback/Dormant: This category should be targeted for re-engagement campaigns so that you can win them back.  

How To use Einstein Engagement Scoring   

Salesforce Marketing Cloud has introduced Einstein Spilt activity within Journey Builder to help Marketers target the specific set of customers with various options as listed below: 

• Persona Spilt- This setting helps you to target customers based on various personas like Loyalist, Window Shoppers, Selective Subscribers, and Win back customers. 
• Web conversion Likelihood Split- This split setting helps you target customers based on their likelihood to convert i.e. make a purchase, download a whitepaper, etc. 
• Click Likelihood Split – This split setting helps you target customers based on their likelihood to click on a link in the email.  
• Subscription Retention Likelihood Split – This split setting helps you target customers based on their likelihood to stay subscribed to your emails.  
• Open Likelihood Split – This split setting helps you target customers based on their likelihood to open the email.  

Below image shows the various settings of the Einstein Split Activity.  

The below image displays Persona Split in the Journey Builder canvas.  

Wrap Up

We’ve listed the various  benefits of the Einstein Engagement Scoring feature and also elaborated on the various use cases for its applicability. We hope the use cases will help you  use Einstein Engagement scoring in the right scenarios.For a detailed understanding of Einstein Engagement Scoring, please visit this link.