Salesforce Data 360, explained: what it is and whether it's useful for you

If your company runs on Salesforce, or is thinking about it, someone has probably mentioned Data 360 by now. Maybe it came up in a meeting with your account team, or a colleague asked “should we be looking at this?” and nobody was quite sure how to answer.

This article is for that moment. It's written for people who don't need to build Data 360 themselves, but who need to understand it well enough to judge whether it fits their organisation, and to follow along comfortably if a project goes ahead. No code, no deep architecture. Just what it is, how it works at a sensible level, what it costs you in effort, and the questions worth asking before anyone commits.

First, a quick word on the name

If you've been reading up and keep bumping into “Data Cloud”, you haven't got the wrong product. Data 360 is simply the new name for Salesforce Data Cloud. The change came in October 2025, when Salesforce launched its Agentforce 360 platform and positioned Data 360 as one of its four building blocks: the unified data layer that gives AI agents the context they need.

In practice, older articles, training material and even some screens inside your own Salesforce org may still say “Data Cloud”. Same thing. We'll stick to Data 360 here.

The problem it's trying to solve

Picture a fairly typical company. Customer and account details live in Salesforce. Online orders sit in an ecommerce platform. Email engagement is in Marketing Cloud. The loyalty scheme runs on something else, and the finance system is the only place that knows who actually paid. Each system holds a piece of the customer. None of them holds the whole picture.

You'll probably recognise some of the symptoms:

• A customer gets a “we miss you” email the day after placing a big order.
• A service agent has no idea the person on the phone is one of your most valuable customers.
• Marketing asks for a list of “customers who bought X but not Y in the last 90 days” and waits weeks for it.
• Two reports on the same question give two different numbers, and a whole meeting goes on deciding which is right.
• Someone wants to try AI, and the first question nobody can answer is “which data would it actually use?”

This is very normal. Salesforce's latest State of Data and Analytics research (surveys run June to August 2025) found the average enterprise uses 897 applications, with only 29% of them connected. Data leaders estimated that 19% of their company's data is siloed or unusable, and incomplete, out-of-date or poor-quality data was still the number one thing stopping organisations from being properly data-driven.

So what is Data 360, in plain terms?

Data 360 is Salesforce's data platform. It pulls together information about your customers (and your business more broadly) from lots of different systems, matches it up into one view of each customer, and makes that view usable inside the Salesforce tools your teams already work in. Salesforce describes it as the data engine native to its platform, feeding workflows, analytics and its AI agents.

An analogy that tends to help: imagine every system in your company is a colleague who knows something about the same customer, but each one keeps their own notes, in their own format, under a slightly different name. Data 360 is the meeting where they finally compare notes. Someone translates so they all mean the same thing by “customer” and “order”, someone works out that “J. Martin” and “Julie Martin” are the same person, and the agreed picture is then handed to whoever needs to act on it.

If you've come across the term customer data platform (CDP), Data 360 is one, and marketing is where it started. Salesforce now positions it more broadly, as a shared data foundation for sales, service, commerce and AI as well as marketing.

A couple of things it is not. It isn't a new CRM; your Sales or Service Cloud stays where it is. And it isn't usually a replacement for a company data warehouse. It's designed to work alongside one, which we'll come to in a moment.

How it works, without the jargon

Under the bonnet there's a lot going on, but for project purposes Data 360 does five things, roughly in this order. These are also the stages you'll hear people talk about in project meetings, so they're worth knowing.

1. Connect. Data comes in from your sources. Salesforce's own apps (Sales, Service, Marketing Cloud and so on) plug in with ready-made connectors, and Salesforce says its connector ecosystem covers more than 200 sources. There's also a second route called zero copy: instead of copying data out of your data warehouse or data lake, Data 360 reads it where it lives, and it works in both directions. That matters if your company has already invested heavily in a warehouse and doesn't want a second copy of everything. Data 360 can also take in unstructured content, like documents and emails, which becomes important when AI comes into the picture.
2. Harmonise. Every system describes things differently. One calls it “Email”, another “cust_email_addr”. In this step, your data is mapped onto a common Salesforce data model, so that “customer”, “order” and “email address” mean the same thing everywhere. It's unglamorous work, and it's where a lot of project time goes.
3. Unify. This is the clever bit, called identity resolution. You set rules for deciding when two records are the same person (for example, same email address and a similar first name), and rules for which value to show when sources disagree (for example, “use the most recently updated address”). Salesforce's Trailhead training compares the result to a key ring: the original records stay intact and linked together, rather than being overwritten by one “master” record. The profile updates itself as source data changes.
4. Analyse and enrich. With unified profiles in place, you can calculate useful things such as lifetime value or days since last purchase (these are called calculated insights), build segments (groups like “lapsed customers in the south-west with an open service case”), and run predictive models.
5. Act. Finally, the results go to work. Insights can appear directly on a contact or account record in Sales or Service Cloud, trigger automated processes in Salesforce Flow, or feed AI agents. Segments can be sent to activation targets, which per Salesforce's Trailhead project include Marketing Cloud, B2C Commerce and advertising platforms such as Amazon Ads.

Running across all five is governance: who can see what, how sensitive fields are tagged and masked, and how consent is respected. Salesforce has been putting more emphasis on policy-based access controls that apply across Data 360, analytics and Agentforce.

What companies actually use it for

The technology is only interesting if it changes something people do on a Monday morning. Salesforce's own Trailhead module on Data 360 use cases groups them by team, and that's a sensible way to think about it:

• Marketing. Build audiences from everything you know (purchases, web behaviour, service history, loyalty status) rather than just email clicks. Then send them to Marketing Cloud journeys or advertising platforms. Just as useful is who you leave out: no promotional email to someone with an open complaint, no “come back” offer to someone who bought yesterday.
• Service. Agents see the bigger picture on the case screen, such as recent orders, loyalty tier and what the customer was looking at on the website, so they don't have to ask questions you already know the answer to.
• Sales. Reps get signals like website engagement or product usage alongside the account record, and companies with several Salesforce orgs can see one combined view of a customer. Trailhead's sales unit also mentions product recommendations and AI-drafted emails grounded in this data.
• AI agents. If you're looking at Agentforce, the agent's answers are only as good as what it can see. Data 360 is the layer that gives an agent customer context and company knowledge (including documents) to base its answers on, under the same access rules as everything else.
• Operations and reporting. Watching key numbers in close to real time, like a spike in service cases or a dip in web conversions, so teams can react the same day.

Where it fits with the rest of Salesforce

The simplest way to picture it: Data 360 sits underneath the Salesforce apps your teams use, rather than being another app beside them. Two connections are worth understanding early:

• Marketing Cloud. This is where it gets a little more nuanced, and it's worth asking about early. With the long-established Marketing Cloud Engagement, Data 360 typically sends it ready-made audiences. With the newer Marketing Cloud Next, Salesforce says Data 360 is built in as the foundation, so the data layer and the marketing tool behave as one product. Which one you have, or are moving to, shapes the project.
• Agentforce. Salesforce's AI agents use Data 360 for customer context and company knowledge. At Dreamforce 2026 in September, Salesforce went further and described Data 360 as one of the core technologies behind what it calls its Enterprise AI Harness, the set of controls meant to make AI agents trustworthy at scale. Some of those new capabilities are planned to start rolling out in early fiscal 2028, which for Salesforce begins in February 2027.

The honest bits

Data 360 can be genuinely useful, but it isn't a switch you flip. These are the things we'd want any client to understand before going further:

• It's priced on usage, not just on seats. Salesforce currently offers three pricing models. Bringing data in (batch loads and zero copy) is free. You pay when you use it: preparing and unifying data, building and sending segments, querying, real-time processing and analysing documents.
• The set-up is real work. Even Salesforce's own Marketing Cloud page acknowledges that data modelling and initial set-up still need investment. Mapping your data to the Salesforce model, agreeing definitions and testing the matching is where most of the project time goes.
• It joins data up. It doesn't make bad data good. Salesforce's Trailhead has a neat example: a sales rep misspelt a customer called Rachel as “Rochelle”, so her sales record didn't match her other records and she ended up with two profiles until the name was fixed at source. Multiply that by thousands of records. Data 360 has tools to tidy and standardise data, but if your source systems are messy, expect some clean-up work alongside the project.
• It needs business owners, not just builders. How strict should matching be? Which system wins when two disagree? Who is allowed to see what? How is consent respected, which in Europe means involving whoever looks after GDPR? These are business decisions, and projects go better when marketing, service and data protection people make them early.
• It's not a shortcut around strategy. If you can't yet say what you'd do with a unified customer view, Data 360 won't tell you. Start with two or three concrete use cases and build from there.

What a Data 360 project actually looks like

Salesforce's own guidance is clear on one point: start with a small, actionable use case that can show value quickly, then grow. In practice, a first project tends to follow these steps:

1. Discovery. Agree the first use case and what success looks like. List the data sources it needs, who owns each one, and how a customer is identified in each (email, customer number, loyalty ID). Salesforce publishes an implementation checklist that covers exactly this.
2. Design. Map the data to the Salesforce data model, agree the matching rules, settle governance and consent, and estimate usage costs.
3. Build. Connect the sources, set up the mappings, matching, calculated insights, segments and activations.
4. Test with the business. This is the step people underestimate. Do the unified profiles look right to the people who know the customers? Do the numbers match what finance or marketing expect?
5. Go live and hand over. Train users, set up monitoring for data refreshes and consumption, and agree who looks after it day to day.
6. Expand. Add the next use case, often reusing much of what was built for the first.

How long all this takes depends mostly on how many sources are involved and how clean they are. A single, well-scoped use case is a very different undertaking from a company-wide customer view.

Who's in the room. Salesforce's Trailhead guidance on building a Data 360 team lists a project manager, a business owner who keeps the work tied to real needs, an enterprise or data architect, the people who configure Data 360, and the end users who'll rely on the results. To that we'd add the owners of each source system, and your data protection lead. You don't need to understand every technical detail, but you will be asked to make decisions, so it helps to know the stages above.

Is it for you? Questions to ask before you start

There's no universal answer, but a pattern comes up again and again. Data 360 tends to earn its place when you already use Salesforce, important customer data sits in other systems, and you can name specific things you'd do with a joined-up view: better-timed marketing, better-informed service, AI agents, or a move to Marketing Cloud Next. If most of your customer data already lives in one place, or your core Salesforce data needs work first, it's usually wiser to start smaller or wait.

To work out which side of that line you're on, answer these honestly:

1. What are the two or three things we would do differently with a joined-up customer view, and how would we measure the difference?
2. Which systems hold the data those use cases need, and who owns each one?
3. How is a customer identified in each system? Is there anything shared, like an email address or customer number?
4. Honestly, how good is that data today?
5. Which Marketing Cloud are we on, or moving to, and how does that change the approach?
6. Do we have a data warehouse, and should zero copy be part of the design?
7. Who will own the matching rules, consent and the usage budget once it's live?
8. Which pricing model suits our use cases, and what does a realistic first year look like in Salesforce's calculator?
9. Could we prove the first use case using the free allowance before committing further?

Data 360 isn't magic, and it isn't for everyone. But for organisations whose customer data is spread across several systems, and who have a clear idea of what they'd do with a single view, it's a practical way to join the dots inside the Salesforce tools their teams already use. If you can answer most of the questions above, you're in a good position to have a productive conversation, whether that's with Salesforce, with a partner or internally.