Heading 1

Heading 2

Heading 3

Heading 4

Heading 5

Heading 6

Lorem ipsum dolor sit amet, consectetur adipiscing elit, sed do eiusmod tempor incididunt ut labore et dolore magna aliqua. Ut enim ad minim veniam, quis nostrud exercitation ullamco laboris nisi ut aliquip ex ea commodo consequat. Duis aute irure dolor in reprehenderit in voluptate velit esse cillum dolore eu fugiat nulla pariatur.

Block quote

Ordered list

Item 1
Item 2
Item 3

Unordered list

Item A
Item B
Item C

Text link

Bold text

Emphasis

^Superscript

_Subscript

The decision on the table is which process gets the first agent. Pick one with fuzzy rules or poor data and the pilot stalls, and the organisation concludes that agents do not work. Pick one with volume and clean data and you get a measurable result. You also get a foundation you can reuse, because the data access, guardrails and monitoring you build for the first agent carry over to the second.

Customer service is the most common first use case

Service desks are where most organisations start, because the work is high in volume and the rules already exist in writing. An agent in Service Cloud can answer questions about order status, returns or delivery times, update the case and hand over to a service employee when a question falls outside its instructions. The value shows up in two numbers: contacts resolved without a person, and handling time for the cases that still reach one.

The limit is the knowledge behind the agent. It answers from your knowledge articles and records, so an outdated article becomes a wrong answer repeated at scale. When we implement Service Cloud, we treat the knowledge base as part of the build, because an agent inherits its quality. Name the person who keeps it current before the agent goes live.

Sales teams gain most from preparation and follow-up

In Sales Cloud, agents earn their place in the work around the deal. They prepare account briefings, summarise calls, draft follow-up emails and flag opportunities that have gone quiet. The sales representative keeps the relationship. That split matters, because a buyer at a large account notices quickly when the conversation is automated.

Lead qualification is a different case. Salesforce also offers agents that qualify inbound leads and book meetings. They suit teams that receive more inbound interest than they can answer. Where your pipeline depends on a small number of large accounts, preparation and follow-up usually return more than automated outreach does.

Marketing and commerce agents depend on unified data

Marketing agents can draft a campaign brief, propose audience segments and write first versions of content, with a marketer approving each step. Commerce agents work on both sides of the webshop. Some help merchandisers run the site, and others help shoppers find a product, check an order or reorder. Both kinds lean on customer data that is complete and current.

That is why many of these use cases run on Data Cloud (Data 360), the Salesforce layer that unifies customer data from every channel into one profile. An agent that recommends a product without knowing the last three orders will recommend the wrong one. The broader case for that layer sits in Data Cloud as a data foundation. An agent that only works with records already in Service Cloud or Sales Cloud can start without it.

Priscilla Calemi

Marketing & Community Manager

How to choose your first Agentforce use case

Score each candidate process on four questions before you commit budget, and compare the scores side by side.

How often the task happens each week, because volume decides whether any saving shows up.
Whether the rules are written down or live in the heads of a few experienced people.
What a wrong answer costs, and whether a person can catch it before the customer does.
Whether the agent can reach the data it needs inside Salesforce or depends on other systems.

A process that scores well on all four is a candidate for the first Agentforce use case. One that fails the last question is a data project first, and it should be planned as one.

Cost depends on usage, and usage depends on design

Agentforce pricing has changed more than once and depends on your contract, so model usage before you look at a price. The drivers are the number of conversations or actions an agent handles, the number of steps each one takes and the data it has to query along the way. Volume grows with success. A use case that works therefore costs more in total while the cost per resolved contact falls.

Measure that against what the task costs today: handling time, cost per contact and time to first response. Then add the internal line most business cases leave out. Someone owns the agent's instructions, reviews a sample of its conversations every week and updates the knowledge it relies on. If the programme stops after one agent, the cleaned knowledge base, the data access and the guardrails keep their value.

Not every task needs an agent

Scoring leads, predicting churn or choosing the best send time are predictive tasks. Einstein features in Sales Cloud and Marketing Cloud handle them without a conversation or an agent. Generative agents are for tasks that need language and action, such as answering a customer or drafting a reply. The difference is set out in predictive and generative AI in Salesforce.

Start with one agent, one owner and one measure

The first agent should go live on one process, with one named owner and one number you report every month. Keep the scope that small. Teams that launch agents in four departments at once rarely learn which design choice made the difference. The groundwork for that first step is covered in first steps with Salesforce AI.

FORWARD's Agentforce accelerator gives that first use case a fixed approach and a clear scope, so you know what the build covers before it starts. When an agent hits a missing feature or a blocking bug, our weekly calls with Salesforce product management put it in front of the people who can fix it.

Explore Agentforce

About the Author

Priscilla Calemi

Marketing & Community Manager

Our insights

Button Text

Salesforce developer working alone beside a Salesforce partner team

Salesforce developer vs Salesforce partner: who to hire when

October 28, 2026 5:08 PM

Agentforce use cases worth starting with, by department

Key takeaways