Audiences Overview Beta

The Audiences feature is in Private Beta, where we work with early users and customers to test new features and get feedback before making them generally available.

Reach out to Customer Success if you are interested in enabling this feature for your workspace.

Audiences lets marketers build targeted customer segments directly from warehouse data and activate them downstream — without writing any SQL. You can define audiences visually in the RudderStack dashboard, keep them in sync as your warehouse data changes, and activate them to multiple destinations on independent schedules.

How audiences work

Audiences are built on top of a data graph, which translates your warehouse data into a business-friendly format that marketers can use to build audiences.

1. Set up a data graph once

Your Data team connects relevant warehouse tables to the Data Graph, mapping them as business entities (for example, Users, Purchases, Accounts), defining relationships between them, and marking timestamped tables as event models.

Click here to view a sample data graph for an ecommerce store

2. Build audiences

Once a data graph is created, Marketers can use the visual Audience Builder to build audiences — by setting segment conditions, and filtering on properties, related records, behavioral events, and users that are part of other audiences.

Click here to view a sample audience named High Value Customers that are not a part of the loyalty program

3. Sync to destinations

Once an audience is saved, marketers can connect it to one or more downstream destinations. Each destination can have its own sync schedule, field mapping, and refresh cadence. Audience membership recalculates at sync time against the latest warehouse data.

Why use audiences?

Building a targeted audience segment — for example, customers who spent over $1,000, purchased 3+ times in the last 90 days, and viewed the pricing page this month — typically requires a marketer to file a ticket with the data team, wait for a custom SQL query, and repeat the cycle for every tweak or new destination.

Audiences removes that bottleneck by letting marketers self-serve within boundaries the data team defines once in the Data Graph.

Who can use audiences

• Growth marketers and campaign managers: Build and iterate on audience segments in minutes instead of filing tickets and waiting days. Combine user attributes, purchase history, and behavioral events, then sync to multiple ad platforms on independent schedules.
• Data team: Define entities and relationships once in the Data Graph. Marketers self-serve within those governed boundaries — no more ad-hoc SQL requests for every segment change.
• Marketing leadership: Reduce time-to-campaign, improve targeting precision with multi-table segments, and free the teams from repetitive audience work.

Key features

• Multi-condition combinations: Combine multiple conditions in a single audience definition without writing any SQL.

• Behavioral event filtering: Filter on timestamped event data with flexible time windows. For example:

  • In the last 30 days
  • Between two dates
  • After a specific date

• Audience inclusion/exclusion: Include or exclude users who are also a part of other audiences. Changes to a referenced audience propagate automatically at next sync.

• Comparison operators: Use comparison operators to create aggregate conditions.

• Multi-destination syncing: Sync a single audience to Meta, Google, LinkedIn, and TikTok, each with its own schedule, field mapping, and refresh cadence.

• Scale and management: Search, filter, sort, and tag audiences for teams managing hundreds of segments.

Common use cases

• High-value customer retargeting: Target customers with high lifetime value who purchased recently and sync the audience to Meta and Google Ads for retargeting campaigns. Adjust thresholds and iterate without involving the data team.
• Churn prevention: Identify high-value customers with no purchases in the last 90 days, exclude anyone recently contacted, and push the segment to an ad platform for a win-back campaign.
• Behavioral segmentation: Build segments based on specific user actions, such as “viewed the pricing page 5+ times in the last 30 days but hasn’t purchased”. Events are first-class entities in the Data Graph, so time-windowed behavioral conditions are point-and-click.
• Suppression and compliance lists: Create a reusable “Recently Contacted” audience once and reference it as an exclusion in every outbound campaign audience. Updates to the suppression list propagate automatically.
• Multi-hop segmentation: Segment users based on attributes two relationships deep. For example, find customers who purchased at flagship stores by traversing a Customers → Sales → Stores path and filtering on store_type = "Flagship".

Get started

To build an audience, you need:

• A Data Graph configured with at least one root entity (for example, Users, Customers, or Accounts). Your data team owns this step.
• A supported destination

See the following guides for more information:

Supported destinations

You can connect an audience to the following destinations:

• Facebook Custom Audience
• Google Ads Remarketing Lists
• LinkedIn Audience
• TikTok Audiences