Marketers / Marketing Cloud

Marketing Cloud Engagement Splits vs. Decision Splits

By Oana Munteanu

Marketing Cloud Journey Builder gives you the ability to bring together multiple communication channels within a single interface. You can create simple or complex customer journeys, to automate marketing communications and guide subscribers through different paths, based on their engagement (or data available within your Marketing Cloud database, or other connected systems).

There are elements available to drag and drop on to the Journey Builder that enable you to control which path subscribers go down, and when. This guide will cover two of the Journey Builder’s “Flow Controls” Marketing Cloud Engagement Splits and Decision Splits.

Journey Builder Flow Controls

Flow Controls are one of the main elements that you’ll find in Journey Builder. Once you configure Flow Controls to your needs, they will decide which paths the contacts will go through. Some of the most commonly used Flow Controls are:

  • Engagement Splits: Looks at the most recent message that was sent from that journey, and checks whether the contact has engaged by measuring specific metrics (e.g. email clicks).
  • Decision Splits: Uses either Journey data or Contact data to send contacts down specific paths in the journey. Which data you should choose will depend on whether data could have changed between the contact entering the journey and the evaluation step.
  • Wait Steps: Hold the contact in that step until the duration of the wait step has lapsed (e.g. 3 days), or until a specific date.

Engagement Splits vs Decision Splits

As we’ve learned this far, both engagement and decision splits are used to control which paths contacts should flow through, based on what they do (engagement data) or what we know about them (contact data). But that’s really all they have in common:

Engagement Splits Decision Splits
DataEngagement dataJourney data or Contact data
Email opens, clicks, bouncesYesNo
Any contact attributes that exist in the journey entry data extensionNoJourney data
Any attributes that do not exist in the entry data extension but are available via Contact Designer NoContact data

Engagement Splits

Engagement Splits look at the most recent message that was sent from that journey (typically the activity that happened before reaching the split).

The flow control checks whether the contact has engaged with the communication by measuring specific metrics such as email opens, clicks or bounces. If they have reached the level of engagement you consider significant, they’ll continue on the ‘Yes’ path, otherwise, they will follow the ‘No’ path.

Note: You can only select one metric per decision split, either opens, clicks, or bounces. When choosing clicks, you can select one (or more) links within one engagement split. Best practice is to give a wait period of at least three days before checking for bounces because the system needs to allow some time re-trying the send before counting it as a bounce, especially when it comes to soft bounces.

Decision Splits

Decision Splits use either Journey data or Contact data to send contacts down specific paths in the journey.

How do you know which data type to use? Think about whether you need to evaluate data that may have changed between the contact entering the journey to the evaluation step.

Let’s use the example of submitting a form to register for an event or purchasing a product. If the answer is ‘yes’ (the contact is now registered or has made a purchase), then you should use Contact data as this data is the most up to date and relevant. If the answer is ‘no’, then Journey data is the right option for optimum system performance, as it’s easier to retrieve this static data than looking up other data sources for no reason.

Note: A single decision split allows you up to 20 paths. These need to be configured using a contact’s attributes (field data) as the split’s evaluation criteria. The rules are evaluated top to bottom, so sequence is important – start with the criteria that you expect most contacts to meet. If you’re familiar with Salesforce lead assignment rules, this works in the same way; a contact that meets more than one set of criteria for a path follows the one they qualify for first.

Do You Want to Experiment?

Path Optimizer is another flow control available in Journey Builder. Create an A/B or multivariate test straight into the journey. Automatically send the winning version to the majority of your subscribers and suppress the less engaging version from being sent out to them.

Einstein-powered Flow Controls

  • Einstein Send Path Optimizer (STO): Einstein (AI) decides the best time to send a message to an individual subscriber based on their previous engagement.
  • Scoring Split: Segments customers based on Einstein Engagement Scoring data and send messages to subscribers that are likely to open them.
  • Frequency Split: Identifies contacts that have been sent too many, or not enough, messages.

Note: Most Marketing Cloud Einstein capabilities need to be activated and your organization needs to have enough data in order for Einstein to work its magic. Then, you can leverage these capabilities with a certain level of confidence.


Remember, Journey Builder is just a tool that supports the delivery of your customer journeys. Engagement Splits and Decision Splits are key elements within Journey Builder that help you to define the most optimized experience for your customers. Therefore, it’s important to understand the main differences between them, such as the type of data evaluated (engagement vs journey/contact data) as well as use cases for each of them (opens/clicks vs registrations/purchases).

Learn more about Journey Builder Flow Controls by checking out the Journey Builder Campaigns module on Trailhead.

The Author

Oana Munteanu

Oana Munteanu is a Salesforce Marketing Cloud Champion and a Senior Marketing Consultant at Bluewolf, an IBM Company.

Leave a Reply