Segment
A Segment in Salesforce Data Cloud is a filter-based audience definition that produces a list of unified Individuals (or other DMO records) matching specified criteria.
Definition
A Segment in Salesforce Data Cloud is a filter-based audience definition that produces a list of unified Individuals (or other DMO records) matching specified criteria. Segments are the bridge between raw unified data and downstream activation: marketers build a segment for "Individuals in California with Lifetime Value over $10,000 who opened an email in the last 30 days" and the same segment activates to Marketing Cloud Engagement, Google Ads, Meta Audiences, or any other connected destination.
Segments are built declaratively in the Segment Canvas, a visual filter builder where users drag DMO attributes onto criteria nodes, combine with AND / OR logic, and preview the audience size in real time. Behind the scenes the segment runs as Data Cloud SQL against the unified profile. Segments materialize on a schedule (typically hourly, daily, or on-demand) and produce a versioned audience snapshot; downstream activations consume the snapshot. The combination of Identity Resolution feeding unified Individuals plus Segmentation producing targeted audiences is the heart of Data Cloud's marketing value proposition.
How Segments turn unified Data Cloud profiles into activatable audiences
The Segment Canvas and declarative filter logic
The Segment Canvas is a node-based visual filter builder. Each criterion node references a DMO attribute (Individual.City, Individual.LifetimeValue, Email Engagement.OpenDate). Multiple criteria combine via AND or OR. Nested groups support complex Boolean logic. The canvas previews the matching audience size as you build, refreshed live, so marketers know immediately if a refinement narrows the audience too much.
Filter by DMO attributes, Calculated Insights, and related entities
Segments can filter on three data types. Direct DMO attributes (Individual.City = California). Calculated Insight values (LifetimeOrderValue > 10000). Related entity exists clauses (Individual has at least one Email Engagement with OpenDate in last 30 days). The combination is what enables sophisticated targeting; static demographics plus behavioural signals plus computed metrics produce far richer audiences than any single layer alone.
Materialization and audience size
Segments materialize on schedule: hourly, every 6 hours, daily, or on-demand. Each run executes the SQL against current Data Cloud data and produces a snapshot. The snapshot is what activations consume. Segments with high churn (engagement-based criteria) benefit from frequent materialization; static segments (demographic-only) can run daily or less. Match cadence to the freshness need; over-materialization wastes credits.
Direct activation versus waterfall and exclusion segments
Three segment patterns. Direct: activate everyone in the segment. Waterfall: activate Tier 1 first, then Tier 2 only for people not in Tier 1, building cascading priority. Exclusion: build an Exclude segment (existing customers, opted-out contacts) and apply it to other segments to remove them. Exclusion segments are how compliance teams enforce do-not-contact lists across every audience without manually editing each one.
Audience preview and the bring-your-own-AI integration
Most segments include a Preview tab showing sample audience members with their key attributes. Newer Einstein Segment Creation features let marketers describe the audience in natural language ("Individuals who bought premium products but have not shopped in 6 months") and Einstein generates the segment criteria. The generated segment is editable; Einstein is a starting point, not the final logic.
Activation: pushing segments to destinations
Once a segment materializes, an Activation publishes the audience plus selected attributes to a destination: Marketing Cloud Engagement, Marketing Cloud Account Engagement, Google Ads Customer Match, Meta Custom Audiences, Amazon Ads, LinkedIn Matched Audiences, or any configured destination. Activations are per-segment-per-destination; one segment can activate to multiple destinations in parallel with different attribute sets.
Segment versioning and audit
Segments are versioned in Data Cloud. Each save produces a new version; activations can pin to a specific version for compliance audits. Segment history shows who edited what and when. Segment size history shows audience-size trends; sudden drops or spikes are flags for ingestion problems, criteria changes, or data-quality regressions.
Building and activating a Segment in Data Cloud
Segment creation runs in three phases: design the filter logic, materialize, then activate to one or more destinations.
- Define the audience criteria
Before opening the Canvas, document the audience in plain language: who is in, who is out, what attributes matter. Pin the business owner to the definition; segment criteria drift fast without an owner.
- Build the segment in Canvas
Data Cloud, Segments, New. Pick the population DMO (typically Individual). Drag criteria nodes onto the canvas. Combine with AND / OR logic. Watch the live audience size as you refine.
- Pick the materialization cadence
Hourly for engagement-heavy criteria, daily for demographic-heavy. Trade freshness against credit cost. Most production segments land between hourly and daily.
- Verify audience size and quality
Compare audience size to expectation. Sample audience members from the Preview tab; verify they match the business criteria. Build a sister exclusion segment if you spot consistent mismatches.
- Activate to destination
Data Cloud, Activations, New. Pick the segment, pick the destination, choose which attributes to send. Schedule the activation cadence. The destination receives the audience plus selected attributes on each materialization cycle.
The base entity the segment filters on. Typically Individual; can be Account, Sales Order, or any DMO. Determines what records the segment outputs.
Hourly, every 6 hours, daily, on-demand. Match to data freshness need and budget.
Marketing Cloud Engagement, Marketing Cloud Account Engagement, Google Ads, Meta, LinkedIn, Amazon Ads, custom REST. One segment activates to many destinations in parallel.
DMO attribute, Calculated Insight value, related entity exists/does-not-exist. Combine via AND / OR with nested grouping.
- Live audience size in the Canvas can lag real materialization on segments with complex joins. Trust the materialized snapshot, not the preview, for final audience size.
- Over-materialization wastes credits. Match cadence to actual freshness need; engagement-driven segments need hourly, demographic-only can run daily.
- Activations are per-segment-per-destination. Changes to segment criteria propagate to every activation using it; coordinate with marketing teams before edits.
- Excluded segments require explicit application; they do not auto-apply across all segments. Build a Do Not Contact exclusion and apply it to every audience that should respect it.
- Einstein-generated segments are a starting point. Review the generated criteria carefully; natural-language interpretation can produce edge-case errors.
Trust & references
Cross-checked against the following references.
- Segments OverviewSalesforce Help
- Create a Segment in Data CloudSalesforce Help
Straight from the source - Salesforce's reference material on Segment.
- Activations OverviewSalesforce Help
- Einstein Segment CreationSalesforce Help
About the Author
Dipojjal Chakrabarti is a B2C Solution Architect with 29 Salesforce certifications and over 13 years in the Salesforce ecosystem. He runs salesforcedictionary.com to help admins, developers, architects, and cert/interview candidates sharpen their fundamentals. More about Dipojjal.
Test your knowledge
Q1. What is a Segment?
Q2. Why is segmentation important?
Q3. How does Data Cloud enhance segments?
Discussion
Loading discussion…