Users of Marketing Cloud have traditionally struggled with segmentation. There was no single, simple way to build and maintain customer segments. Users combined formula fields, custom fields, Campaigns, Flows, record types, reports, and sometimes Apex. Each method had its own limits. As needs grew, these workarounds became hard to scale. Users often had to revisit their approach.

Even after creating segments, users struggled to validate results and identify segment members. Keeping segments updated as customer data changed was also hard. 

That’s because ExactTarget, which was renamed to Marketing Cloud, was originally built as an enterprise engine structured around relational database tables, not marketer-friendly user profiles.

Today, by shifting the core data layer to Data Cloud, Salesforce separated data storage from segmentation logic, making real-time, visual segment building viable across vast datasets.

That’s what we’ll talk about today. 

Our chief interest is to dismantle the pseudo-divide between segmentation in Data Cloud and in Marketing Cloud Engagement. Let’s kick off.

Segmentation in Data Cloud vs Marketing Cloud Engagement 

Before going into the details, consider this guiding principle: use Data Cloud for durable, company-wide definitions and Marketing Cloud Engagement for tactical, campaign-specific audiences.

If an audience requires a consistent definition across all teams and channels, such as “active customer,” “high-value,” or “churn risk,” define it once in Data Cloud and reuse it. Audiences created for a single send or journey, based on behavior within SFMC, should be managed in SFMC.

Now, let’s draw closer to the picture

Data Cloud segmentation vs Marketing Cloud Engagement is not the right way to frame it.  

Marketing Cloud Engagement offers two distinct segmentation tools:

• Filtered data extensions are quick and user-friendly, but they only filter a single source data extension and cannot join across data extensions or data views. 
• SQL queries in Automation Studio can join across data extensions and Marketing Cloud data views, which is necessary for tasks involving send, open, click, or journey history. 

But that also means you require strong SQL skills. 

Data Cloud segments, on the other hand, operate across all connected sources, use unified person-level profiles instead of subscriber records, support real-time use cases, and can activate to multiple destinations. They also consume credits and are typically managed by a separate team. 

When to use Data Cloud for segmentation

Data Cloud is appropriate when Marketing Cloud Engagement cannot meet your audience requirements. Use Data Cloud in these scenarios:

• Your data resides outside Marketing Cloud Engagement (such as commerce, service, offline, web, or warehouse sources), since the platform can only segment data it contains.
• You need person-level, deduplicated audiences instead of subscriber-level ones.
• You require activation to multiple destinations.
• You need a durable, company-wide audience definition.
• You require real-time and streaming triggers rather than scheduled batches.

However, Data Cloud is not suitable in four common scenarios, which are often overlooked. Avoid using Data Cloud in these situations:

• The data resides in Marketing Cloud Engagement and does not leave it, as this adds unnecessary cost and latency.
• You need a one-off, disposable audience, since Data Cloud creates a durable asset for a single use.
• Send-context logic (such as suppressions, exclusions, or send-time rules), which should remain within the send process.
• The audience is based on Marketing Cloud Engagement’s own engagement data (such as opens, clicks, or journey behavior); rebuilding this data upstream is redundant.

When to use Marketing Cloud Engagement for segmentation 

A common scenario is when teams need openers or clickers from a specific journey. This data is generated and stored in Marketing Cloud Engagement’s data views and will be used for a send within the same platform. Creating a Data Cloud segment for this purpose requires waiting for another team and incurring additional costs, even though a SQL query can provide the same results directly. 

The same principle applies to joins across data views, campaign-specific audiences that marketing should build independently, and any audience used exclusively for sends within Marketing Cloud. 

Keep in mind that arguing for Marketing Cloud on the basis of speed can be risky. When teams build audiences independently without clear guidelines, the same concept may be defined differently, leading to inconsistencies that only become apparent when reports conflict during a QBR. 

Use speed as a tiebreaker for tactical audiences.

Don’t use it to justify redefining company-wide concepts at a local level. 

What about billing 

In Data Cloud, segmentation and activation are billable. So, recreating an audience that Marketing Cloud Engagement can handle natively will use up credits without providing additional value.

In Marketing Cloud Engagement, the billing impact is more significant and less widely understood. 

A record is counted as a billable contact as soon as it reaches a journey’s entry source, even before the entry filter is applied. Therefore, activating an overly broad Data Cloud segment in Marketing Cloud Engagement can increase your billable contact count, including records that never enter the journey. 

To avoid unnecessary costs, refine your audience in the segment, rather than relying on the entry filter. 

Some segments can be published once and used as-is, avoiding unnecessary costs for recurring refresh cycles. Decide on refresh frequency for each segment based on its specific use case, rather than applying a uniform schedule. Refresh frequency directly affects credit consumption.

Summing up

Framing this as a binary choice is misleading. The two systems are intended to work together. Activating Data Cloud for Marketing Cloud Engagement creates a sendable data extension in SFMC. The established approach is to split responsibilities: Data Cloud defines the audience, while Marketing Cloud Engagement manages execution, including suppressions, splits, and personalization at send time.