09.09.26 By Scott Effler
We’re heading into Dreamforce this year with a different goal than usual. We’re not chasing every “here’s what’s new” headline. We’re going in with a running list of “here’s what we still don’t know.”
The hype is impossible to miss: headless architecture, Claude integrations, Salesforce showing up inside Slack, Teams, and a dozen other tools you already have open. The main idea across all of it is the same, meet people where they already work, instead of asking them to come to you.
That idea is compelling. So, instead of a session preview, we’re looking at the questions driving our conversations at Dreamforce and the platform shift still taking shape. These are the things that will tell us whether this shift is as far along as the keynotes suggest, or whether we’re still earlier than the momentum implies.
Over two decades, companies have layered in third-party apps, custom logic, and internal processes that assume the Lightning and even Visualforce interface is where the work happens. That history is an asset, but it’s also friction.
The emerging Headless Experience Layer (HXL) promises to expose that same logic in new surfaces i.e., Slack, Teams, ITSM tools, custom-built apps, without forcing anyone back into Salesforce itself. That means the underlying components need to shift from Lightning to more portable, React- or Angular-based building blocks. It’s happening, but unevenly. Components that simply display information; cards, dashboards, or summaries are maturing quickly. Components that let you act; approve, update, and execute are further behind.
That gap matters more than it sounds. Seeing an update is convenience. Acting on it without switching tools is the actual promise of headless. So, my questions for Dreamforce are, “What can I, and can’t I, do today?” “How will Salesforce help customers and partners to enable this technology?” “What will be the promise of this technology as we roll into Dreamforce 2027?”
Every platform shift needs an early win, a use case valuable enough to matter and simple enough to actually ship. We’ve seen this pattern before with AI adoption generally: the fastest wins come from narrow, well-defined tasks, not sweeping transformations.
The early candidates for Salesforce’s version of this seems to be Slack integrations, Claude-powered workflows, and exposed skills that let other tools call Salesforce logic directly. These make sense on paper. Whether they hold up in practice is a different question.
Here’s the uncomfortable possibility: the biggest blocker to headless Salesforce might not be technical at all. It might be organizational habit.
Consider that plenty of companies, more than a decade after Lightning launched, are still running Salesforce Classic. Migrating processes, retraining teams, and unwinding years of built-up process builder workflows is slow, disruptive work, and most organizations don’t prioritize it until they’re forced to.
If a UI shift within the same platform took over a decade to reach full adoption for some customers, a shift that asks teams to work in entirely different tools, while trusting that Salesforce logic still runs correctly behind the scenes, is a much bigger ask. The technology can be ready long before the organization is.
So, the question we want answered is how Salesforce plans to bring both ends of its customer base along together: the companies eager to move fast, and the companies still finishing their last migration.
While at Dreamforce, we want to hear where real adoption is running ahead of expectations, and where it’s stalling. Our guess, based on how these transitions usually go, is that technical debt will keep coming up as the real constraint. Most customers aren’t short on enthusiasm for new capabilities, they’re boxed in by what they’ve already built, integrated, and depend on running correctly.
So, the question for Dreamforce is simple: does the sentiment we hear directly from customers match the story being told from the stage? If there’s a gap, that gap tells us more than either version on its own.
The most compelling version of this vision goes something like this: instead of doing the work, then remembering to update Salesforce afterward, the update happens naturally as a byproduct of the work itself. Close a deal in conversation with Claude, and Salesforce updates automatically, no separate data-entry step, no dropped follow-up.
That’s a real improvement if it works as described, and it doesn’t stop with the person who owns the deal. Execution and delivery teams often live in Slack, Teams, or Jira rarely opening Salesforce directly. If closing a deal can automatically trigger onboarding steps or hand off execution details into the tools those teams already use, that’s not just a UI convenience, it removes an entire category of manual handoff and the errors that may come with it.
We’ve seen a version of this work at a smaller scale already. A global risk assessment firm we worked with had a strong Salesforce foundation, but reps were still losing time digging up account and risk intelligence before every customer conversation. Surfacing that intelligence directly inside the workflows reps were already using, instead of as a separate research step, cut sales prep time by up to 50% and proposal rework by up to 30%. It’s a small proof point for a much bigger idea: the question is how much of “work from wherever you are” is already live in production today versus just being a demo built for a keynote stage. Those are very different maturity levels, and the difference matters for anyone deciding whether to plan around this now or wait a cycle.
We believe the questions we have will be top of mind to our clients as they evaluate what is ready today, what remains uncertain, and where the greatest opportunities lie. This is also very much the lens behind our own Growth OS work this year: the belief that real value shows up when data, digital products, and the platforms underneath them work as one connected system instead of a pile of point solutions. Headless Salesforce is a live test of exactly that kind of unification, which is why we’re watching it so closely.
If you’re heading to Dreamforce too, we’d love to hear what you’re watching for. And if any of this overlaps with what you’re working through, headless timelines, Salesforce technical debt, or just figuring out what’s real versus what’s “safe harbor”, we’d welcome the chance to grab a few minutes while we’re all there.
Bridgenext is a certified Salesforce Partner and a member of Salesforce’s Forward Deployed Engineering (FDE) Partner Network, a small group recognized specifically for delivering Agentforce in production, not just in pilot. That work, combined with over 15 years in the Salesforce ecosystem, is what makes these questions practical for us rather than academic. Check out where you can find us at Dreamforce if you’d like to connect.
