Serviceblazer Recap: 5 Key Insights from Dreamforce 26

Dreamforce is always an exciting experience, but with so many announcements, innovations, and conversations happening at once, it can also be a lot to process. As a Serviceblazer Community Group Leader, I wanted to bring our community together after Dreamforce ’26 to look beyond the announcements and explore what these developments‌ mean for the work we do every day.

More importantly, I wanted to create an opportunity for Serviceblazers to share their perspectives, ask questions, and have an open conversation about both the opportunities and challenges ahead.

I was lucky to be joined by three Serviceblazers who live this work every day:

• Christina Nava, Senior Manager of Salesforce & RevOps at Stored 
• Jeanette Jett, Senior Principal Consultant and Team Lead at InvestorFlow
• Janeen Marquardt, Partner and Chief Technical Strategist at Maple Digital Transformation 

Here’s what came out of the conversation, and what to check out if you missed this year’s Dreamforce.

1. Salesforce is becoming the backbone of Service

One standout from this year’s event: the transformation of Salesforce’s interface through AIforce. Your data, business rules, permissions, and automations can power your service experience wherever the work happens. Maybe you work in the Service Console. Maybe you work in Slack or an AI interface. 

But this shift represents more than a change in how users interact with Salesforce. It reflects a broader evolution in the platform’s architecture and the role it plays within an organization.

Salesforce is no longer defined solely by the interface we log into. It’s becoming the foundation that powers business processes across multiple channels, providing the data, automation, security, and business logic needed to support those interactions. As more work happens outside the traditional Salesforce interface, the importance of strong data architecture, well-defined permission models, reliable automation, and effective governance only increases.

For Salesforce professionals, this shift has significant implications. The architectural decisions we make today will increasingly determine not only how people interact with Salesforce, but also how AI agents access information, execute processes, and operate within established guardrails. Getting these foundations right is essential to ensuring that both human users and AI agents can work effectively, securely, and consistently, regardless of where those interactions take place.

Check out this announcement from the Service Cloud keynote:

2. The Service Console isn’t going away

I’ll admit this was the uncomfortable question: If reps spend less time in the Salesforce UI, were all those years of optimizing record pages and layouts wasted? The panel’s answer was no. Why? Simple tasks will move to where reps already work, but complex cases will still live in the console.

There’s comfort in knowing the information is there at our fingertips, without having to write a new prompt on the fly. The console becomes a go-to place that works alongside your AI agents.

3. Admins are becoming architects

Salesforce administrators have long played an important role in data architecture, security, automation, and business process design. I think the shift we’re seeing with AI is less about administrators becoming architects and more about these existing architectural responsibilities becoming increasingly critical.

As AI agents begin interacting with our systems, the decisions we make about data structures, permissions, and automation have a much greater impact.

“You can’t just be somebody who’s an order taker and just do the thing that’s being asked for. You’ve got to understand the bigger picture and the impacts.” — Janeen Marquardt

The traits the panel kept returning to:

• A willingness to learn. 
• Asking for help. 
• Asking why. 

None of us can call ourselves experts in a world changing this fast. Ready to learn yourself? I suggest starting with Trailhead content on data architecture and Flow best practices.

4. Deploy AI agents deliberately

Our group discussed whether we felt ready to put AI service agents in front of customers today, and our answers differed. The sticking point was the use case. We’re ready to deploy agents to answer questions grounded in good knowledge articles and handle low-risk reversible actions. Industries like financial services, however, should exercise more caution and run thorough pilots before deployment.

While we’re still learning how AI-specific risks manifest in real-world implementations, the fundamentals of responsible technology adoption remain the same. Strong governance, appropriate permissions, thorough testing, and continuous monitoring are essential. As our experience with AI grows, so will our ability to anticipate challenges and establish effective safeguards.

Want to learn more? Check out The Trust Map: Guardrails, Governance, and Control You Can See on Salesforce+.

5. The human job gets more focused

Having started my career in customer service, this question hits particularly close to home. I remember what it was like to be the person answering those calls, navigating difficult conversations, and trying to resolve problems while meeting performance expectations.

As AI takes on more routine interactions, human representatives may increasingly find themselves handling situations that require greater judgment, empathy, and problem-solving skills. That raises an important question: Are organizations prepared to evolve their expectations alongside the technology?

If representatives are handling a greater proportion of complex or emotionally demanding interactions, traditional performance metrics may no longer tell the full story. We need to consider whether we’re providing the right training, support, and measures of success for this changing environment.

The technology may change how service is delivered, but we can’t lose sight of the people delivering it.

It’s an interesting topic among Serviceblazers at the moment. From what I gather, it comes down to making sure we’re using the advances in technology to help the real people on both sides of a service call.

Conversation starters for your team

One of my favorite things about the Salesforce community is that we all bring different experiences and perspectives to the table. While we may not always agree on the best approach, those differences often lead to the most valuable conversations. That’s exactly what I hoped to accomplish with this meetup. 

Looking to keep the convo going with your own team huddle? Here are five questions from our session:

1. What would we let an AI agent do without human approval, and what would we never‌ do?
2. What does our team do with the time freed up from simple requests?
3. Where do our case conversations live: on the record, in Slack, or in DMs?
4. Who can build automations or agents in our org, and should they?
5. What KPIs would we add if AI joined our service team tomorrow?

As Salesforce continues to evolve, having opportunities to ask questions, challenge assumptions, and learn from one another becomes even more important. You don’t need to have all the answers to contribute something meaningful to the conversation. I want to invite you to join the Serviceblazer Community on Slack. It’s a free-to-join, central meetup location for service pros to share knowledge and celebrate each other’s wins. See you there.