The entire Salesforce ecosystem is shifting from scheduled batch processing to event-driven architectures. For data architects and marketers, this shift means you no longer have to cross your fingers and hope your nightly data jobs finish in time for the morning email run. By combining Real-Time Pipelines with Data Cloud-Triggered Flows, you can activate data while the customer is still engaged with your brand.
Key Points: Real-Time Pipelines & Triggered Flows
- Streaming Ingestion: Data arrives continuously via the Ingestion API rather than waiting for a scheduled bulk upload.
- Instant Harmonization: As data streams in, it is immediately harmonized into a unified customer profile.
- Data Cloud-Triggered Flows: The ultimate game-changer. When a segment publishes or a Data Model Object (DMO) updates, Salesforce automatically fires a Flow.
- Actionable Events: That Flow can instantly send a Marketing Cloud journey, ping an external API, create a Case for a service agent, or notify a sales rep in Slack.
A customer adds a $500 item to their cart but abandons it.
In a batch world: The audience segment refreshes overnight. The reminder email arrives the next morning—by which time the customer has already bought the item from a competitor.
In a real-time world: Using streaming ingestion and a Data Cloud-Triggered Flow, the segment updates in near real-time. The Flow fires, pushing the user into a Marketing Cloud journey, and they receive a 10% discount text message within 15 minutes while they are still browsing their phone.
Frequently Asked Questions
A nightly batch cannot meet a minutes-level SLA (Service Level Agreement). To fix this, you must change the pipeline:
- Use the Ingestion API so the basket event streams into Data Cloud immediately.
- Ensure the segment is configured for near real-time publishing.
- Build a Data Cloud-Triggered Flow that listens for that segment publish and automatically fires the outbound message via Marketing Cloud or a third-party API.
Developer Tip: Always ask the business if they truly need "minutes." If a 30-minute delay is acceptable, you can design a much cheaper, less intensive solution.
Volume. A segment publish can qualify tens of thousands of people at the exact same moment. If your Flow makes an individual HTTP callout for each person, you will immediately exhaust your Salesforce API limits or overwhelm the receiving system. Your Flows must be bulk-safe, safe to re-run, and have strict entry guardrails to prevent infinite loops (where updating a record triggers the same segment to publish again).
This is a data mapping issue. The browse events were likely mapped to the Other category instead of the Engagement category in Data Cloud. Only objects mapped as "Engagement" carry the specific time-series functionality required to build duration-based segments.
You are missing a semantic layer. If you rely on the BI tool (like Tableau or CRM Analytics) to calculate revenue on the fly, different dashboard builders will use different formulas. You need to use Calculated Insights in Data Cloud to define the metric exactly once. Then, every dashboard simply reads that pre-calculated number.
Data Cloud processes massive volumes of information. While it is incredibly fast, it is not perfectly instant. Always use the phrase "near real-time" when setting expectations with stakeholders or interviewing for a role. Promising literal "instantaneous" execution proves to senior architects that you haven't actually run streaming pipelines in a live enterprise production environment.
- The Rule: Batch is based on a schedule. Real-time is based on an event. Data Cloud-Triggered flows convert data changes into actionable events.
- The Gain: Massively shortened time-to-value. You can market to, or service, a customer while they are actively engaged.
- The Price: Data Cloud runs on consumption billing (Data Services Credits). Streaming data and continuous processing burn through credits faster than overnight batches.
- The Limits: Segment publishes can process massive audiences at once. Ensure your downstream systems (and your Flows) are prepared to handle sudden spikes in volume.