For the past year, Agentforce has been the most talked-about capability in the Salesforce ecosystem. It’s also been one of the most misunderstood. A lot of business leaders have heard the name, seen it in Salesforce marketing, and still aren’t sure what it does in practice or whether it’s relevant to their business right now.

The Summer ’26 release answers part of that question. Agentforce is now generally available inside Salesforce setup — the part of the platform where administrators configure how the system works. That’s a specific, concrete capability that’s ready to use today, and it’s worth understanding what it actually means before your team starts exploring it.

What Salesforce Setup Is and Why It Matters

If you’ve never spent time in the administrative side of Salesforce, setup is essentially the control panel for the entire platform. It’s where fields get created, permissions get assigned, workflows get configured, and objects get built. It’s the part of Salesforce that most business owners never see, but it’s where the structure of your entire environment lives.

Traditionally, everything in setup has required a human administrator to click through menus, fill out forms, and manually configure each element. For a simple change — adding a field to a record, adjusting a permission — that might take a few minutes. For complex configuration work, it can take hours or days.

Agentforce in setup changes that by allowing an AI agent to assist with configuration tasks using natural language. Instead of navigating through menus, an administrator can describe what they need, and the agent can help execute it.

What Agentforce in Setup Can Do

The specific capabilities that are generally available now include tasks like granting permissions, creating fields, and assisting with configuration changes that would otherwise require manual navigation through the setup interface. The agent understands the structure of Salesforce, knows what’s possible within setup, and can take action based on plain-language instructions.

This isn’t a chatbot that gives answers. It’s an agent that can do things. The distinction matters because it changes what your administrator’s time gets spent on. Instead of clicking through repetitive configuration steps, they can describe the outcome they want and let the agent handle the mechanical work.

The result, in practice, is that more of your administrator’s time goes toward thinking strategically about how your Salesforce environment should work rather than executing the steps to make it happen.

The Credit Situation Worth Understanding

One thing to be aware of before your team starts using Agentforce in setup: while setup actions don’t consume the standard Agentforce credits that most AI features draw from, they do use a small amount of Data Cloud credits.

Salesforce has indicated in their documentation that they plan to eliminate this Data Cloud credit consumption, but that change hasn’t gone into effect yet. For most businesses, the consumption is minimal enough that it’s not a financial concern. But if your team is actively using Agentforce in setup at scale, or if your Data Cloud credit allocation is limited, it’s worth monitoring until Salesforce makes the promised adjustment.

Your Salesforce administrator can track this through the product limit tracking dashboard that’s also new in this release — a single view that shows utilization across all the consumable elements of your Salesforce contract.

The Agent Analytics Dashboard

The second meaningful Agentforce update in this release is a new performance dashboard that Salesforce is calling the Agent Analytics Dashboard. This is where you go to see what your agents are actually doing.

Before this dashboard existed, measuring the impact of Agentforce was largely qualitative. You could sense that things were moving faster or that your team was spending less time on certain tasks, but you couldn’t quantify it in a way that held up to scrutiny.

The dashboard changes that. You can see how often agents are being used, what types of tasks they’re handling, and how their performance is trending over time. For a business leader who has invested in Agentforce or is considering doing so, this is the tool that lets you answer the question every AI investment eventually has to answer: is this actually working?

What This Signals About Where Agentforce Is Going

The availability of Agentforce in setup and the addition of performance tracking are both signals of the same thing: Agentforce is moving from pilot territory into the core infrastructure of the platform.

Early AI features tend to be optional add-ons that enthusiastic teams experiment with while the rest of the business waits to see if they’re real. What Salesforce is building now is different. Agentforce is being woven into how the platform functions — into setup, into performance monitoring, into the tools administrators use every day.

That trajectory matters for business leaders because it means the question is shifting. It’s no longer “should we try Agentforce?” It’s “how are we going to use it, and how will we measure whether it’s working?” The businesses that start building answers to those questions now are going to have a meaningful head start over the ones that wait.

What to Do With This Information

If your business is on Salesforce and you haven’t had a conversation with your administrator about Agentforce in the last few months, that conversation is overdue. Specifically, it’s worth understanding whether Agentforce in setup is enabled in your environment, whether your administrator has started using it, and what the plan is for measuring its impact through the Agent Analytics Dashboard.

You don’t need to have a fully developed AI strategy before that conversation happens. You just need to start asking the questions that will shape one.

If you’re not sure where to begin or what Agentforce in setup would actually look like for your specific environment, reach out using the link in the show notes.

Listen to the full podcast episode here.

Related Resources

• What Agentforce Actually Is — A Plain-Language Guide for Business Owners
• Connecting Claude and ChatGPT to Your Salesforce Data — What Salesforce’s MCP Servers Make Possible
• Salesforce Flow Updates That Actually Reduce Errors and Save Your Team Time
• The Salesforce Summer ’26 Release: What Business Leaders Actually Need to Know