Salesforce Data Cloud Pricing 2026: What Does Customer 360 Really Cost?
The honest DACH guide to Salesforce Data Cloud, today Data 360, and Agentforce costs. With credit model, typical cost scenarios, hidden costs, and a realistic ROI framework for Austrian companies.
The first question in these projects is rarely technical.
Usually, a team sits in the kick-off and says, in so many words: we have 50 sales users, want cleaner segments, a better customer picture, and maybe test Agentforce first in service. Do we need to budget 120,000 euros per year or rather 250,000?
That's exactly where Salesforce pricing gets difficult. Not because it's deliberately opaque. But because today license, consumption, storage, agent logic, and implementation costs all come together.
One additional point that gets confused in many older articles: Salesforce renamed Data Cloud to Data 360 effective October 14, 2025. At the same time, Salesforce now communicates the model as a flexible, consumption-oriented setup around Consumption Credits, Storage, and Add-ons. Anyone still working with pricing comparisons from 2024 or early 2025 articles is often comparing terminology that is no longer used in current conversations.
This guide isn't selling a glossy business case. It's giving Austrian companies a realistic framework for budgeting a Data 360 or Customer 360 initiative.
Why Data 360 Pricing Is So Often Confusing
The old Data Cloud model had multiple parallel counters: Unified Profiles (how many customer profiles are merged), Activations (how often data is exported for marketing or other target systems), and different credit types for different operations.
The new model is somewhat clearer, but not automatically simple. At the core, three cost blocks come together that should be understood separately: Consumption Credits for active usage such as data ingestion, queries, scoring, or Agentforce conversations, Data Storage for stored volume, and Premium Add-ons for additional functions such as Identity Resolution or specific AI and agent functions. This is exactly how Salesforce also communicates the Data 360 pricing logic today.
The typical mistake happens very early in the project. On paper, the base license still looks reasonable. Only in the second step does it become clear that the real lever lies in usage, storage, and additional features. In other words: the platform often looks cheaper or more expensive at the start than it then actually is in real operations.
Why Agentforce Is Its Own Cost Block
Agentforce should not simply be run under "AI" in the budget. The model is too independent for that.
According to current Salesforce communications, AI in sales is no longer a niche topic: 87% of sales organizations already use some form of AI. At the same time, 54% of surveyed sellers say they have already worked with agents. This is not just a trend slide for sales. It explains very well why Agentforce suddenly lands on the table in so many budget and architecture conversations today.
With Agentforce, you're not simply unlocking a feature and then running everything unlimited. Instead, multiple consumption logics interact. Agentforce Conversations are billed per autonomous conversation — a complete interaction segment from query through response to completion or escalation.
Then there are Agentforce Actions. These are the moments when an agent actually does something, such as updating a record, triggering an action, or accessing knowledge. In the official Salesforce rate cards, these operational steps run under Flex Credits. In parallel, there are Einstein Requests for generative AI processing at the LLM level. Put simply: one measures thinking, the other measures doing.
It gets interesting when the agent additionally accesses Data 360 context, such as a complete customer profile or purchase history. Then consumption credits also accrue on the Data 360 side in parallel. Practically speaking: a single Agentforce interaction can draw credits from two pots. This is regularly overlooked in early calculations.
Typical Cost Scenarios for DACH Companies
Important upfront: Salesforce does not publish classic public list prices for Data 360 like it does for standard user licenses. The following figures are therefore indicative values from typical project calculations, partner experience, and publicly available benchmarks. What ultimately appears in the contract depends heavily on volume, term, and negotiation.
Scenario A: Mid-market with 2,500 active customers and 50 sales users
In a typical mid-market setup, we often see a combination of marketing segmentation, basic customer profiling, and a rather simple Agentforce chat integration. This is not a huge enterprise project, but also far from a small CRM sandbox.
For Sales Cloud Enterprise with 50 users, a realistic annual range is usually €60,000 to €80,000. A Data 360 base license typically runs €25,000 to €40,000. For Agentforce Basic with around 5,000 conversations per month, we typically see €12,000 to €20,000 per year. Adding Marketing Cloud Account Engagement (Pardot) usually puts the range at €20,000 to €35,000.
This puts pure annual license costs at roughly €117,000 to €175,000. On top come one-time implementation and configuration costs of €30,000 to €60,000 and annually €15,000 to €25,000 for managed services and support.
Scenario B: Enterprise with 50,000+ records and 200+ users
In an enterprise context, it's usually about a complete Customer 360 picture, active Agentforce in the contact center, and tightly integrated marketing automation. Technically this is significantly more demanding, and organizationally usually also.
For Sales Cloud plus Service Cloud Unlimited with 200 users, you should realistically budget €300,000 to €450,000 per year. Data 360 Enterprise is typically €80,000 to €150,000. For Agentforce Enterprise with around 50,000+ conversations per month, we roughly see €60,000 to €120,000. If a MuleSoft integration layer is also needed, that often adds another €80,000 to €120,000.
On the license side alone, such a setup therefore often lands at €520,000 to €840,000 per year. For implementation and migration, you should budget one-time €150,000 to €350,000. Managed services and support in such setups is often €60,000 to €100,000 annually.
These numbers are deliberately to be read as indicative values, not as an offer. In practice, there's almost always room for negotiation, especially with multi-year contracts and larger package volumes.
What Appears Too Late in Many Budget Plans
In almost every Salesforce project, there's a moment when someone realizes that the license is expensive, but not the real problem. The real problem is usually that around the license there's a whole wreath of costs that were only half-considered in the initial calculation.
The first major block is implementation effort. Data 360 is not a switch you flip and then all customer data is cleanly merged. It involves data modeling, source connections, mapping, testing, and often architecture decisions that can't easily be reversed later. As a rough guideline, we typically see 30 to 60% of annual license costs as a one-time implementation investment.
The second point is change management and training. Salesforce projects in reality fail far less often due to the platform than due to adoption. If teams don't accept new processes, if data isn't maintained cleanly, or if Agentforce use cases technically work but nobody knows how to use them meaningfully in daily life, the best license won't help. Typically we're talking 8 to 15% of total project costs.
Another classic is data quality. Data 360 is only as good as the data flowing into it. If the existing CRM has duplicates, inconsistent formats, or missing required fields, it won't automatically become a clean Customer 360 model. Then, before the actual project, a data cleansing project often first emerges.
Then there's custom development. Many companies start hoping that standard functionality will suffice. In practice, specific approvals, special logics, Flow customizations, or Apex developments quickly appear. The more individualized the processes, the larger this block becomes.
Finally, maintenance and upgrades shouldn't be forgotten. Salesforce releases multiple times per year and custom developments or integrations need to be checked after larger releases. This is not a one-time expense, but ongoing operations.
Why the 3-Year View Is Almost Always More Honest
Salesforce is still often treated like an annual purchase: contract, license, project start, done. For economic evaluation, that's usually too short-sighted.
A more useful perspective is the 3-year TCO. Then you can see much more clearly how licenses, implementation, operations, and further development relate to each other.
In the first year, licenses are at 100%, one-time implementation at 40 to 60%, managed services at 15%, and training at 8%. In sum, that lands at 163 to 183% of the annual license.
In the second year, licenses usually stay at 100 to 110%, because usage and package scope often grow slightly. For implementation or further development, another 5 to 10% accrues, managed services stays at 15%, and training is often 3%. Total: 123 to 138% of the annual license.
In the third year, we often see 100 to 120% license costs, another 5 to 10% for adjustments, 15% managed services, and 3% training. That totals 123 to 148% of the annual license.
Calculated over three years, a Data 360 project realistically lands at around 454% of annual license costs, or approximately 4.5 times the annual license. That's exactly why looking at the Year 1 price alone leads in the wrong direction in many cases.
When the Investment Still Pays Off
The good news is: a cleanly set up Salesforce and Data 360 project can be economically very worthwhile. But not because "Customer 360" sounds good — because concrete processes run better.
An example with deliberately transparent assumptions: If a company processes 20,000 Tier 1 tickets per year, 50% can be automated, and a manually processed ticket costs an average of €25, that yields a potential of €250,000 per year. This is not a universal market figure, but a calculation based on clear assumptions.
With Customer Lifetime Value, we see a different effect. If Data 360 is actually used for personalized communication and better cross-sell logic, 10 to 15% CLV increase as an industry benchmark is a realistic reference value. That doesn't happen automatically, of course, but it shows the order of magnitude the benefit can reach.
Similarly with churn reduction. If predictive analytics help identify at-risk customers earlier and retain them, even 1% less churn can make a noticeable difference. With a customer base of €10 million ARR, that's then €100,000 in additional revenue.
Time-to-Quote is also often more relevant than first appears. When CRM, proposal, and approval processes are cleanly integrated, we often see 30 to 40% time savings from lead to proposal in projects. It doesn't appear in any license line item, but in sales it quickly makes a measurable difference.
What's Often Underestimated in Negotiations
Salesforce prices are negotiable — in many cases even more so than buyers initially assume. Whoever enters a budget conversation should not only know the product names, but also their own usage logic.
A 3-year commitment is usually cheaper than an annual contract. That only pays off if usage is somewhat predictable, but that's often where the best price lever emerges.
Bundling also makes a difference. Whoever negotiates Sales Cloud, Service Cloud, Data 360, and Agentforce together sits in a different position than someone buying individual products one after another. That's not a secret trick — it's simply sales logic.
Timing is also relevant. Salesforce's fiscal year ends on January 31. Around this date and at quarter-end, the incentives in sales are often particularly high to still close deals. Whoever negotiates at these points therefore often has more room to maneuver.
And quite practically: credit packages should be calculated in advance. Especially with Agentforce, it makes a big difference whether 5,000 conversations per month are realistic or whether it ends up being 15,000 instead. Whoever understands this logic early negotiates more cleanly and plans more realistically.
What We Usually Recommend to Our Customers
As an Austrian Salesforce partner, we see relatively often the same patterns. Some companies invest too little in the wrong place; others too much in the wrong place.
The most common mistake on one side is starting with only a Sales Cloud license and adding Data 360 later. That sounds sensible, but often results in data modeling and integration logic having to be rethought later.
On the other side, we see companies buying very large Agentforce packages very early, even though it's not yet clear which use cases will really generate volume. Then the budget is there, but productive usage isn't yet.
Our recommendation is therefore usually pretty unspectacular:
-
Start with one clearly defined use case, not a platform vision.
-
Measure credits and usage early, instead of just estimating them.
-
Then scale based on data — only grow bigger when behavior, volume, and value are really visible.
Exactly at this point, a sober external view often pays off the most. Not to make Salesforce seem smaller or larger, but to bring the investment to a size that fits the company.
FAQ
Has Salesforce Data Cloud changed prices in 2026?
Salesforce renamed Data Cloud to Data 360 effective October 14, 2025 and has since communicated a simplified, consumption-oriented model around Consumption Credits, Data Storage, and Premium Add-ons. Existing contracts generally continue until contract expiry; renewals are then transitioned to the current structure.
Can I use Agentforce without Data Cloud or Data 360?
Yes, basic Agentforce functions are also possible without Data 360. But the real added value usually only emerges when the agent can access a more complete customer profile, purchase histories, or other consolidated data.
What's the difference between Einstein Requests and Agentforce Actions?
Einstein Requests refer to the use of generative AI at the LLM level. They arise when Salesforce processes a prompt or AI request through the LLM gateway. Consumption depends on, among other things, the model type and the scope of the request.
Agentforce Actions, on the other hand, are executive steps of an agent on the platform — such as updating records, triggering a flow, or retrieving knowledge. These are measured in the official rate cards via Flex Credits. Put simply: Einstein Requests primarily measure AI processing, Agentforce Actions primarily measure the agent's operational actions.
Conclusion
Anyone talking about Salesforce Data Cloud, today Data 360, shouldn't just look at a license number. What matters is the interplay of credits, storage, Agentforce usage, implementation, and ongoing operations.
The annual license is important, but it's rarely the whole truth. A budget only becomes really reliable when use case, data quality, integration effort, and adoption are honestly factored in.
And that's exactly where the difference usually lies between a project that looks good on paper and a setup that actually works in practice.