Salesforce Data Cloud (formerly Customer Data Platform / Genie) — hyperscale data layer for unified customer profiles across systems.
Capabilities:
- Ingestion from many systems (Salesforce CRM, Marketing Cloud, ERP, web, mobile).
- Identity Resolution — match records representing same person.
- Calculated Insights — aggregates and computed fields.
- Activation — push unified data back to operational systems and ad platforms.
- Segmentation — natural-language and rule-based for marketing.
When you need Data Cloud:
- Multi-system customer view — data scattered.
- Marketing personalisation at scale.
- AI / ML that needs unified data.
- Real-time experiences based on broad customer context.
When you don't:
- Single Salesforce org with everything in one place.
- Simple use case without unified profile need.
- Cost-sensitive — Data Cloud is significant investment.
Architecture integration:
1. Data flows in.
- Salesforce CDC -> Data Cloud.
- Marketing Cloud -> Data Cloud.
- External via APIs or batch.
2. Identity Resolution.
- Configure rules: email match, phone match, fuzzy name+address.
- Output: unified profile per real person.
3. Calculated Insights.
- Lifetime value.
- Engagement score.
- Last touch.
- Custom metrics.
4. Activation.
- Push to Sales Cloud / Service Cloud as fields.
- Push to Marketing Cloud as audiences.
- Push to ad platforms.
- Real-time triggers.
5. AI / ML.
- Einstein and Agentforce use Data Cloud as data foundation.
- ML models trained on unified profiles.
Architecture decisions:
- Data Cloud as MDM vs MDM tool feeding Data Cloud.
- Real-time vs batch ingestion.
- Activation paths — which downstream systems?
- Segmentation strategy — who creates segments?
Talent:
- Data Cloud architects emerging specialty.
- Data engineering skills important.
- Salesforce + data platform combo skill set.
Common pitfalls:
- Data Cloud as solution looking for problem — without clear use case, expensive.
- Underestimating data engineering — significant work.
- Identity Resolution accuracy — false positives/negatives common.
- Cost surprises — usage-based pricing can scale unexpectedly.
Senior architect insight: Data Cloud is strategic infrastructure for Salesforce-centric AI/personalisation. Required for advanced Agentforce / Einstein use cases. Worth investing in early if you're heading there.
The senior framing: Data Cloud is to data what Salesforce is to CRM — opinionated, integrated, premium-priced. Choose deliberately.