Getting Started with Pardot Einstein Behavior Scoring

Pardot Einstein brings together AI and Machine learning functionality to Pardot users. It allows users to become more efficient and focus on the right kind of prospects while making data-driven decisions to alter your marketing strategy to be more effective based on what’s working for you. Einstein analyzes data from both Pardot and Salesforce to show scores and insights on your most engaged prospect with the most effective campaigns. This result helps to align marketing and sales with the right information so both parties can be invested and involved. Here’s how to get started with Pardot Einstein Behavior Scoring!

The Traditional Pardot Scoring Model
Let’s start with the basics, if you’re a Pardot user you’re likely familiar with the prospect activity table. This table displays the types of engagement a prospect has taken with your organization. This can be anything from an email open to a form conversion to a custom link click. You can check out a complete prospect activity glossary here. These activities have been previously used to feed the traditional prospect scoring model in Pardot, but recently there has been a lack of confidence with that model. Are form completions really a higher value than let’s say an event registration? Are all website page visits really weighted equally? The traditional scoring model also has a major flaw with its lack of consideration around timing. Without depreciating the score the score can keep accumulating over time. So you can have an inactive prospect showing the same score as an active prospect when in reality we all know those two prospects should be treated the same. The traditional model doesn’t consider patterns that prospects have taken, and how more valuable actions be weighted. The challenge here is making scoring more efficient, introducing Einstein Behavior Scoring.

What is Einstein Behavior Scoring?
It’s not a new concept but a data-driven approach to solving an age-old problem, how to accurately identify prospects whose behavior suggests that they are ready to buy. Einstein Behavior scoring uses a machine learning approach to prospect scoring leveraging the Pardot prospect activities that show what people are doing when they’re engaging with us. What Einstein does it looks at all of the positive activities it took for a person to get added to an opportunity. Einstein will start to see some patterns, and apply different weights and scores to those different activities and touchpoints depending on when they took place. For instance, when we consider the value of time, if someone completes a form today, it will weigh more than someone who filled out the same form a month ago. It’s no longer just an action, we can now add time into consideration. From there we can more accurately try to figure out what we should do with them, should they be sent to sales, should they be nurtured, have they completed the appropriate activities to go in either direction.

How does Einstein Actually work?
The behavior score is more intuitive than the previously used traditional Pardot score. It gives you insights to let you know some of the top activities. It provides insight as to why someone has a certain score. It will even go down to a specific form or piece of content. This is really helpful information for the sales team to leverage during the sales calls. Additionally, the behavior score takes it one step further and provides the top negative actions. So if a lead is becoming less engaged both marketing and sales will have visibility and can take action.

What does Einstein Score Actually Predict?
The Einstein Behavior score helps identify prospects that are going to or are likely to generate an opportunity. It will predict how likely a prospect is to generate an opportunity compared to all other prospects and based on how they’re engaging with your marketing efforts.

How Can I Use This in ABM
If you haven’t noticed most of the scoring is on the individual person level, not the account. If you’re using ABM you’ll likely want to see the behavior score at the account level. You can accomplish this by creating a standard report for accounts, and group the report by account. Then take a look at what the maximum behavior score in all of your accounts. This should give you insight into which accounts have the highest behavior scores.

Einstein Lead Scoring, What’s the Difference?
Einstein’s behavior scoring is focused on marketing engagement. The Lead Scoring component is based on the fields on the lead. And how does it compare to the other leads we have converted, It’s more of a fit score, whereas behavior is more focused on behavior. Einstein’s lead score is all about the fields of the leads, and what we know about them. So we can take the attributes of a person, such as their title, are they a decision-maker, how large is their company, and use this information to create an analysis of whether or not they are a good fit. So it’s almost like their Pardot grade. The Behavior score is more about the leads engagement with your marketing. So the combination of these two scores is pretty powerful to see whether or not they are a good fit for your organization and whether or not they are interested based on the behavior score.

How does Behavior Score Actually Work and Evolve?
Salesforce is pulling in up to 1 year of prospect engagement history data from Pardot. Using a trained model, it delivers behavior scores and key factors. It automatically learns as patterns change, so every 10 days, a new model is trained. It will be able to take into account new Pardot assets, new marketing changes, and prospect activity changes. Einstein reallocates the weights and establishes new patterns every 10 days.

There is no data pooling or sharing, the information and patterns are established based on your individual org. This way the patterns can learn off of your individual data and produce scores and weights based on your model and information only.

How to Get Started with Einstein Behavior Scoring
First, you need to go to the Pardot Einstein features in Salesforce, and enable them via the setup menu. Navigate to setup > search for Einstein Pardot > then turn on the feature > edit your page layouts in the Lightning App Builder to decide where you want to place your new Einstein features.

Using Einstein Behavior Score with Pardot
A new field will appear in Pardot between the current Pardot score and grade. It looks like a standard field, but it is actually created as a prospect custom field, so when you’re looking to use it in automation, be sure to check the custom field list. It’s called Einstein Behavior Score. So you should see a score populated for any prospect who has had an activity in the last year. All prospects will be scored, not just prospects connected to a lead/contact in Salesforce. With the new field, you can now use this field in Pardot logic such as automation rules, adding them to an engagement studio nurture program, and other automation. A great example is when you have prospects that aren’t ready to go to sales, you can use the behavior score to nurture them until they are ready to go to sales.

Einstein Behavior Scoring Dashboard in B2BMA
There is an out of the box dashboard that helps you understand what Einstein is uncovering, you’ll:

• Understand the strength of influence each engagement signal has on your scoring.
• Discover which marketing assets are influencing conversions.
• Take action and drive sales alignment directly from the connected prospects list.

FAQ for Pardot Einstein Behavior Scoring

Do I need a special license to implement Einstein Behavior Score?
Currently, it is included in Pardot Advanced, so if you have that edition you automatically get access to all of the Pardot Einstein features. You also get business units, so you can separate your business units into accounts. You also need to have Salesforce Enterprise edition or above, mostly because of API restrictions.

How is Einstein Behavior Score Difference from Prediction Builder?
Prediction builder is more if a general-purpose solution that allows you to focus on a Salesforce object, and say ‘I want to predict this in this field’. Whereas behavior scoring is meant to solve the age-old problem of prospect prioritization and understand their engagement. The solution is built in a way that automatically knows how to pull data.

I am New to Pardot & Don’t Have Any Prospect Activity
For new customers, who don’t have historical data in your org, there is a baseline model of Einstein Behavior scoring you can leverage. Salesforce recommends at least 6 months of activity, and 20 prospects attached to an opportunity in order to generate a more accurate model to provide behavior scoring. It’s not something that you chose, but the baseline model will be used if there isn’t enough information. It pulls from the rules-based model. Once you have enough information, it will automatically change from the baseline model.

Why Can’t I See Scores in Pardot?
If you have manually chosen a connect user to Pardot, you need to assign a permission set to the connect user. They will need a Sales User, CRM User, or Service User permission set. This is only if you manually chose your connector user and not using the default connect user.

If you’re ready to harness the power of Einstein Behavior scoring just let us know, we’re happy to help you get the most out of Pardot. 

About Roycon
We’re an Austin-based Salesforce Consulting Partner, with a passion and belief that the Salesforce platform’s capabilities can help businesses run more efficiently and effectively. Whether you are just getting started with Salesforce or looking to realize its full potential, Roycon specializes in Salesforce Implementations, Salesforce Ongoing Support, and Salesforce Integrations, and Development. We’re the certified partner to guide the way to increase Salesforce Adoption, make strategic decisions, and build your Salesforce Roadmap for success.