EMINENCE VSP Talk to the architect

The Salesforce partner built for banking — lending journeys live this quarter, not eighteen-month programmes.

AI · DATA· 6 min read · by Srinivasa, Founder & Architect · published Feb 2026

What is Data 360 for banks? The intelligence layer under your agents, explained

Salesforce renamed Data Cloud to Data 360 and every bank's diagram gained a box. Here is what the box actually does, why it isn't a data warehouse, and when a bank genuinely needs it.

What is Data 360?

Data 360 — Data Cloud until late 2025 — is Salesforce’s data layer: it connects the systems where your customer data actually lives, resolves the fragments into unified profiles, and makes the result usable inside the platform — by workflows, by analytics, and increasingly by AI agents that need grounded answers. For a bank, the headline is the architecture: much of this happens zero-copy, reading your warehouse or core banking data where it sits rather than building another copy to govern.

What it is not

It is not a replacement for your data warehouse, not a core banking system, and not a magic deduplicator that absolves you of data ownership. Your warehouse remains the analytical home; Data 360 is the operational bridge that lets the platform — and an agent answering a customer at 9:31 on a Tuesday — reach the right fact with the right permissions, live.

What Data 360 actually does for a bank

Connects without copying. Zero-copy connections read core balances and warehouse history in place — no second copy of sensitive data, no second governance regime, no overnight sync pretending to be real time.
Resolves identity. The customer who exists in five systems under three spellings becomes one unified profile, with match rules you control and can defend — the difference between an agent answering from a record and answering from the record.
Computes live facts. Calculated insights — balances aggregated across products, household exposure, churn signals — become attributes the workflow and the agent can use in the moment.
Feeds AI with permissions intact. Agentforce grounds its answers through this layer, inheriting the platform’s security model — the agent sees what the requesting user may see, no more.

When does a bank actually need it?

Honest test: if your use case is one journey on data already inside FSC, you may not need Data 360 on day one — and a partner who says so is worth keeping. The moment it earns its keep is when the journey needs facts that live outside — core balances in a decision, warehouse history under an agent, one customer view across brands. Buy it for the use case that needs it, sized to that use case, not as an act of architectural faith.

An agent is only as honest as the data layer underneath it. Data 360 is where that honesty is built — or isn’t.

The honest caveat

Identity resolution is policy, not software: the match rules that merge two records embody a risk decision your bank must own. Over-merge and you show one customer another’s data; under-merge and the agent answers from a fragment. We treat ruleset design as a governance workshop with named sign-off — the configuration is the easy part.

Three questions before the diagram gains the box

Which use case needs data the platform can’t already reach? Name it, or defer the licence.
Who signs off the match rules? Identity resolution is a risk policy wearing a config screen.
Zero-copy or copy — per source? Every copy is a second governance regime you’re volunteering for.

How Eminence VSP helps

We implement Data 360 for banks the unglamorous way: scoped to the first use case, zero-copy where the source allows it, match rules designed as governed policy — the foundation work our Agentforce readiness article describes. See AI & Agents or talk to the architect.

S.
Srinivasa
Founder & Architect, Eminence VSP — the person who scopes and delivers these builds.
Share on LinkedIn →

Let’s talk.

One conversation with the architect — and a clear view of what your bank could ship next quarter.

Talk to the architect