Most companies need one core cloud before anything else: Sales Cloud when the pain is pipeline and forecasting, Service Cloud when the pain is cases and response times. Every other product, from Marketing Cloud to Data 360 and Agentforce, should wait until that core is adopted and its data is trusted. Pick by the job a product does, not by its feature list. This chooser covers each product, a master table, and a buying order that keeps licenses from going unused.
How do you narrow down the Salesforce product list?
Start with a business problem and the team that owns it, then find the product built for that job. A product without a named owner and a measurable problem is a product to defer.
Salesforce sells many products on one shared platform. Its product page lists Agentforce, Data 360, Sales Cloud, Revenue Cloud, Service Cloud and Field Service and Operations. Marketing Cloud, Commerce Cloud, Tableau, MuleSoft and Slack follow. Because they share one platform, a contact or account created by sales is the same record service and marketing see.
Names shift often. Data Cloud became Data 360, Field Service now appears as Field Service and Operations, and AI features are sold as Agentforce add-ons such as Agentforce for Sales. Treat any name in this guide as current at the time of writing, and check it against your quote.
Is a cloud the same thing as an edition?
No. A cloud is a product, such as Sales Cloud; an edition is the tier of that product you license, such as Enterprise or Unlimited on older contracts, or Core, Advanced and Max since September 2026.
The edition sets limits and included features: API calls, sandboxes, storage and which automation tools you get. Two companies can both own Sales Cloud and have very different capabilities because of edition. Licenses are then assigned per user, and a user can hold more than one. Our guide to Salesforce license types explains how user licenses, permission set licenses and add-ons stack. Edition contents change between releases, so confirm specifics with your Salesforce account team.
Which core cloud should come first?
Buy the cloud that matches the team losing the most time or revenue today. For most B2B companies that is Sales Cloud; for support-heavy businesses it is Service Cloud.
- Sales Cloud. Job: leads, accounts, opportunities, forecasting and pipeline reporting. You need it when deals live in spreadsheets and nobody trusts the forecast. Wait if you have one or two sellers and a simple funnel. Pairs with Account Engagement, Revenue Cloud and Slack.
- Service Cloud. Job: cases, queues, routing, entitlements and a knowledge base. You need it when requests arrive by email and phone with no tracking or SLA reporting. Wait if a shared inbox still handles volume cleanly. Pairs with Experience Cloud, Field Service and Agentforce.
Plenty of companies run both in a single org. Our Sales Cloud vs. Service Cloud comparison covers that decision in depth, so this guide does not repeat it.
Which add-ons extend sales and service work?
Field Service, Experience Cloud and Revenue Cloud each extend a core cloud rather than replace it. Add one when the core process works but stops at a clear boundary.
- Field Service (listed as Field Service and Operations). Job: work orders, scheduling, dispatch and a mobile app for technicians. Signs you need it: dispatchers juggle calendars by hand and techs arrive without parts or job history. Not yet: a handful of jobs a day that a calendar handles fine. Pairs with Service Cloud.
- Experience Cloud. Job: portals and sites where customers, partners or members log in to Salesforce data. Signs you need it: staff rekey partner deal registrations or answer status questions customers could look up. Not yet: no clear audience or nobody to own content. Pairs with Service Cloud or Sales Cloud.
- Revenue Cloud (which replaced Salesforce CPQ for new customers). Job: product catalogs, pricing rules, discount approvals, quotes and contracts. Signs you need it: complex bundles, frequent pricing errors or slow approval chains. Not yet: a short price list where standard Quotes cover it. Pairs with Sales Cloud and often an ERP.
Salesforce now presents Revenue Cloud as its quoting and revenue lifecycle product. If you already run Salesforce CPQ, ask your account team where it sits on the roadmap before expanding it.
Marketing Cloud Engagement or Account Engagement: which marketing product fits?
Account Engagement, formerly Pardot, suits B2B teams nurturing leads into a sales pipeline. Marketing Cloud Engagement suits high-volume, consumer-facing messaging across email, mobile and journeys.
- Account Engagement. Signs you need it: marketing hands sales cold leads and cannot show which campaigns produced pipeline. Not yet: no agreed lead definition between sales and marketing. Pairs with Sales Cloud.
- Marketing Cloud Engagement. Signs you need it: large consumer lists, segmented journeys and several channels. Not yet: a small list and a single monthly newsletter. Pairs with Data 360 and Commerce Cloud.
Salesforce also markets Marketing Cloud Next as its newer flagship marketing product. How it is packaged alongside the existing two is still evolving, so get the current lineup from your account team.
When do Data 360 and Agentforce earn a place?
Both come after the core cloud works. They amplify good data and clear processes, and they amplify bad ones too.
- Data 360 (formerly Data Cloud). Job: unify customer data from many systems into profiles and segments that other products can act on. Signs you need it: customer data spread across commerce, product, support and marketing tools. Not yet: one CRM with duplicate problems you have not fixed. Pairs with Marketing Cloud and Agentforce.
- Agentforce. Job: AI agents that answer, route and act inside Salesforce, sold as add-ons such as Agentforce for Service and Agentforce for Sales. Signs you need it: high volumes of repeatable requests with documented answers. Not yet: thin knowledge articles, messy case data or no owner for agent behavior. Pairs with Service Cloud, Knowledge and Data 360.
Our separate Data 360 guide covers readiness in detail. The short test is whether your current CRM data is clean enough to trust today.
Do you need Tableau, CRM Analytics, MuleSoft or Slack?
These are cross-cutting tools. Each solves a problem that appears once several systems or teams are involved.
- Tableau or CRM Analytics. Job: analysis beyond standard reports, across large or blended datasets. Signs you need it: leaders export Salesforce data to spreadsheets to answer basic questions. Not yet: standard dashboards are unbuilt or ignored. CRM Analytics lives inside Salesforce; Tableau, including the newer Tableau Next, reaches wider data sources.
- MuleSoft. Job: an integration layer connecting Salesforce with ERP, billing, data warehouses and legacy systems. Signs you need it: many point-to-point connections that break when one system changes. Not yet: one or two simple integrations a native connector already handles. Pairs with any cloud.
- Slack. Job: collaboration around Salesforce records, alerts and approvals. Signs you need it: deal and case conversations scattered across email threads. Not yet: your company standardized on another chat tool and has no appetite to switch. Pairs with Sales Cloud and Service Cloud.
Should you buy an industry cloud instead of a core cloud?
Sometimes. An industry cloud adds a data model and processes for one sector, but it still depends on a solid core configuration underneath.
Salesforce currently markets Financial Services Cloud, Health Cloud, Manufacturing Cloud, Nonprofit Cloud, Education Cloud and Consumer Goods Cloud, among others. Some arrive as add-ons layered over Sales or Service Cloud licenses, and others as their own licenses. That packaging varies and changes, so confirm it with your account team before you budget.
- Financial Services Cloud: households, financial accounts and advisor relationships for wealth, banking and insurance teams.
- Health Cloud: patient or member records, care plans and care team coordination.
- Manufacturing Cloud: sales agreements, account forecasting and channel partner programs.
- Nonprofit Cloud: fundraising, programs, volunteers and outcomes tracking.
- Education Cloud: recruitment, admissions, student success and advancement.
- Consumer Goods Cloud: retail execution, store visits and trade promotion work.
The test: would you otherwise rebuild that sector's data model from scratch with custom objects? If yes, the industry cloud usually pays off. If your process looks like generic B2B selling, a core cloud is often enough. Our guides to Financial Services Cloud and Health Cloud go deeper on those two.
Which product matches which business need?
Use this table as a first pass. The right column matters most: it lists the conditions that mean you should hold off.
| Business need | Product | When to wait |
|---|---|---|
| Track deals and trust the forecast | Sales Cloud | One or two sellers with a simple funnel |
| Track, route and measure support requests | Service Cloud | A shared inbox still copes with volume |
| Schedule and dispatch technicians | Field Service and Operations | Few daily jobs and no parts tracking |
| Give customers or partners a login | Experience Cloud | No audience or content owner defined |
| Quote complex products and pricing | Revenue Cloud | Short price list; standard Quotes suffice |
| Nurture B2B leads into pipeline | Account Engagement | Sales and marketing disagree on a qualified lead |
| Send high-volume consumer journeys | Marketing Cloud Engagement | Small list and one channel |
| Unify customer data across systems | Data 360 | Core CRM data still duplicated or stale |
| Automate repeatable requests with AI | Agentforce | Knowledge and case data not yet reliable |
| Analyze large or blended datasets | Tableau or CRM Analytics | Standard dashboards not yet in use |
| Connect many systems reliably | MuleSoft | One or two simple integrations |
| Collaborate on records in chat | Slack | Another chat tool is firmly in place |
| Sector-specific data model | An industry cloud | Your process is generic B2B selling |
What order should you buy Salesforce products in?
Buy in layers: core cloud, then the extensions that core users ask for, then data and AI. Each layer should be adopted before the next is signed.
- Layer one: the core cloud for your most painful process, configured well and adopted by the team that owns it.
- Layer two: extensions the core team keeps hitting walls without, such as quoting, portals, field scheduling or marketing automation.
- Layer three: cross-system tools like MuleSoft and analytics, once several systems need to agree.
- Layer four: Data 360 and Agentforce, once your records are clean and your processes are written down.
An industry cloud can move into layer one if its data model fits from day one. Retrofitting one later often means remapping records you already migrated. A partner who works across clouds can help you map this order on a roadmap before you sign.
How do you keep Salesforce licenses from becoming shelfware?
Only license what a named team will use soon, and check usage before every renewal. Shelfware usually starts with a bundle bought for a future project that never got staffed.
- Tie each product to an owner, a problem and a target metric before signing.
- Ask for implementation capacity, internal or partner, in the same budget as the license.
- Review login history and feature usage once go-live settles, and again before renewal.
- Be wary of multi-product bundles where only one product has an active project.
- Write down which products you declined, and why, so the next renewal starts from facts.
