Data Cloud Foundations
What Data Cloud is
Data Cloud is a hyperscale customer data platform (CDP) built on Salesforce's Hyperforce infrastructure. Its data store is separate from core Salesforce objects: records ingested into Data Cloud live in Data Cloud objects, not in Account/Contact tables, so "Data Cloud writes records into standard CRM objects by default" is always a wrong answer.
The pipeline you must be able to recite
Every domain on this exam hangs off one chain: Data Stream → Data Lake Object (DLO) → field mapping → Data Model Object (DMO) → Identity Resolution → Unified Individual → Insights/Segments → Activation. The DLO is the raw, source-shaped landing copy. Mapping harmonizes it into DMOs that follow a shared schema (Individual, Contact Point Email, Contact Point Phone, Contact Point Address, Party Identification, Engagement, and custom DMOs). Segments and calculated insights query DMOs and unified profile objects, never raw DLOs.
Organizing concepts
Data spaces partition data, metadata, and access - a user's data space determines what they can see. Data kits package data streams, DMOs, mappings, and related metadata for deployment between orgs. Standard DMOs model people/parties, contact points, engagements, and transactions; custom DMOs extend them when no standard object fits.
Traps
- Confusing DLO with DMO, or assuming mapping is optional.
- Assuming identity resolution is automatic - a unified profile only exists after a ruleset runs.
- Assuming Salesforce sharing rules, roles, or org-wide defaults govern Data Cloud access.
- Forgetting that a data space assignment affects the whole pipeline for that user.
Rule of thumb
If a field is visible in the data stream but missing from a segment, walk the chain backwards: mapped to the DMO? Related correctly? Included in the ruleset's match fields? The failure is almost always mapping or relationship, not segmentation.