Salesforce is infamous for its rebrands. 

So much so that it arouses users by the mere fact of taking place, whenever it does take place yet another time, resuming the vicious cycle of having to re-emphasize its apparent pointlessness. Even with the unspoken agreement among users to stick to legacy names, a new name triggers the assumption that the change must be anchored in some underlying product evolution, however minor. 

The confusion around the Salesforce Data Cloud rebrand follows a similar pattern.  

But behind the smoke and mirrors of rebrands, it’s clear that the data layer is the true foundation of Salesforce’s agentic agenda. In today’s untangling exercise, we are going to look past the changing labels and clear the mist on this critical foundation. Let’s get started.

What is Salesforce Data 360?

At Dreamforce 2025, Salesforce renamed Data Cloud to Data 360. It marked the platform’s sixth name, following Customer 360 Audiences, Salesforce CDP, Marketing Cloud Customer Data Platform, Salesforce Genie, and Salesforce Data Cloud. Importantly, the product, license, data model, and integrations remain unchanged. Existing configurations will continue to function as before.

The change is in positioning. “Data Cloud” emphasized data centralization, while “Data 360” presents the platform as a live context layer supporting AI, automation, and decision-making. 

It is now formally part of the Agentforce 360 suite as its data foundation.

Salesforce Data 360 is a layer, not a feature 

Data 360 serves as an intermediary layer between your data sources and the applications that use this data. It integrates systems of record such as CRM, ERP, web and mobile platforms, commerce, data warehouses, file stores, and legacy applications. It provides unified, reliable data to applications including Sales, Service, Marketing, Commerce, Tableau, Einstein, and Agentforce.

Remember, Data 360 is not a standalone tool for isolated tasks; it is infrastructure that connects to data wherever it resides. It enables direct querying of platforms such as Snowflake, BigQuery, and Databricks through zero-copy access, eliminating the need for ETL or duplication. 

This approach ensures a consistent data view across all applications.

How data moves through Salesforce Data 360

Within the Data Cloud layer, data progresses through five stages. 

• It is ingested via real-time data streams or batch processes and stored as Data Lake Objects.
• Next, raw objects are mapped to the standardized Customer 360 Data Model, creating Data Model Objects (DMOs). 
• Identity resolution then links records belonging to the same individual into unified profiles. 
• The data is further enriched with calculated insights and data graphs. 
• Finally, it is activated and published to segments, flows, agents, and external destinations. 

Data Spaces provide logical partitions, separating data by brand, region, or business unit without requiring a separate organization.

Salesforce Data 360 doesn’t replace enterprise data platforms, ETL tools, middleware, MDM, or traditional CDPs. Instead, it acts as a customer-focused operational layer that consumes curated data from those systems and makes it actionable inside Salesforce. 

What is a Unified Profile?

In Data Cloud, a Unified Profile is a single, comprehensive view of a customer created by merging data from many different systems. It is created through identity resolution. 

The platform applies match rules to identify records referring to the same individual across systems, then uses reconciliation rules to combine their attributes into a single profile. This process connects a customer’s multiple contact points, such as emails, phone numbers, and addresses, to one identity while retaining full source lineage. 

This approach does not create a “golden record.” Data 360 uses a key ring model, preserving and linking source records rather than overwriting or merging them. The unified profile remains mutable, updating continuously as source data changes, new sources are added, or resolution rules are adjusted. 

This design ensures the profile stays accurate and auditable, rather than locking in a single master version that may become outdated.

What are Data Graphs?

A data graph is not a unified profile or a chart; it is a performance structure. 

Data graphs transform normalized DMO table data into materialized views within a data space. 

In practice, a data graph is a pre-flattened record organized by a primary key, such as a unified individual. It can be visualized as a two-column table: one column contains the unified individual ID, and the other contains a deeply nested JSON object with all related, pre-joined data.

Data graphs exist to improve speed. When customer data spans many related objects, live joins on each request are too slow for the sub-second response times required by AI agents. With a data graph, the agent retrieves a single pre-built record rather than executing multiple queries.

This capability enables Agentforce to provide unified data quickly enough to support agents.

Salesforce Data Cloud rebrand: Long story short

Agentforce agents require live, accurate customer context to be effective. This context is delivered through unified profiles and data graphs, generated by Data 360. 

Salesforce’s product leadership had positioned 2025 as a pivotal year for AI and data, and renaming the data platform to anchor Agentforce 360 reflects this commitment.

The key takeaway is that agents are only as good as the data they’re acting on. 

Most enterprise AI failures stem from data issues. If you plan to leverage Agentforce, you are inherently adopting Data 360, whether viewed as a CDP or AI initiative. 

Therefore, the data foundation warrants the same level of scrutiny as the agents.