BA, Admin, or Dev: Your Path to Full Stack Salesforce
The previous piece made the case that the Admin/Developer divide is collapsing — and that the role replacing it, Full Stack Salesforce, combines platform depth, business analysis, and stakeholder management into one job.
The question that follows is obvious: how do you get there from where you are?
The answer is different depending on your starting point. And the most important thing to understand upfront: AI closes most of the technical gaps. The gaps it can't close are the ones that take the longest to build.
If you're coming from BA
You have the most undervalued starting point of the three. You already understand the business — the priorities, the politics, the way decisions actually get made. You know how to work a room of stakeholders and translate what they want into something documentable. That's not a soft skill. That's the hardest part of the job, and most technical people can't do it.
Your gap is implementation. You know what needs to happen but not how to make it happen inside Salesforce. Historically that meant learning flows, learning Apex, spending months building technical knowledge before you could bridge the two worlds.
AI changes that equation significantly. You can now describe a business process in plain English and use AI to generate the flow logic, the Apex class, the integration pattern. You still need to understand what you're looking at — you can't blindly deploy AI-generated code — but the barrier to getting your hands on a working implementation has dropped dramatically.
The move: get hands-on in a Salesforce org, lean on AI to help you build things, and let your existing business knowledge guide what you're building and why. The Admin certification is your fastest formal entry point.
If you're coming from Admin
Most admins are strong on process and weak on logic. Not because they aren't capable — but because nobody ever made them sit down and think through conditionals, loops, and error handling. The click-through nature of Salesforce tools lets you build automations without ever confronting the underlying logic. Until the automation breaks in a way you can't explain.
That logic-building muscle is what needs to develop. Not necessarily coding — though coding helps — but the ability to think through a problem systematically: if this, then that, unless this other thing, handle this edge case. That's the foundation of both good flows and good code.
AI is useful here too. Not to write the logic for you, but to explain it. Walk through a complex flow with an AI, ask it why each step is structured the way it is, ask what breaks if you remove a condition. Use it as a thinking partner while you build the mental model. The goal isn't to become a developer — it's to stop being afraid of complexity.
The move: pick one real automation project that's slightly beyond your comfort zone. Build it. Break it. Fix it. Repeat until the logic stops feeling foreign.
If you're coming from Dev
This is the most important path to get right — and the hardest.
Here's the uncomfortable truth: AI can already write most of the code that Salesforce developers write. Apex classes, LWC components, SOQL queries, integration handlers — all of it is within reach of a well-prompted AI tool today, and that capability is only going to improve. The technical moat that made "I'm a Salesforce developer" a defensible position is eroding.
What AI cannot do is walk into a room with a CFO who has decided the project is going in the wrong direction and change their mind. It cannot manage a stakeholder who is resistant to change because the new system threatens their team's headcount. It cannot earn the trust of a business that has been burned by a failed implementation before. Those things require presence, judgment, and a track record — and they are built over years, not prompts.
For developers, the gap to close isn't technical. It's human. Stakeholder management, communication, and the ability to translate between what the business wants and what the platform can actually deliver — these are the skills that will define who wins in a world where AI handles the code.
The move: stop sitting in the technical corner of every project. Start attending discovery sessions. Volunteer to run a stakeholder meeting. Write the summary email after the workshop. Put yourself in the room where the business decisions happen, not just the ones where the technical decisions happen. It will feel uncomfortable. That discomfort is the work.
Every path has a gap. AI closes most of the technical ones faster than anything that's come before. The gaps it can't close — the human ones — are exactly where you should be investing your time.
That's the actual syllabus for getting to Full Stack Salesforce. Not more certifications. The work that can't be automated.
Frequently Asked Questions
What exactly does 'Full Stack Salesforce' mean, and is it an official Salesforce role title?
Full Stack Salesforce is not an official Salesforce job title but a description of a practitioner who combines platform configuration, development, business analysis, and stakeholder management — replacing the older split between Admin and Developer tracks. Companies increasingly want one person who can own a Salesforce implementation end-to-end rather than hand work between siloed specialists. The value is less about any single technical skill and more about the ability to connect business requirements directly to working solutions.
If AI can generate Apex and flow logic from plain English descriptions, does a Full Stack Salesforce professional still need to learn those tools deeply?
Yes — the ability to evaluate, debug, and take responsibility for AI-generated output requires genuine understanding of the underlying platform. Deploying code or automation you can't explain introduces serious risk in a production org, particularly around governor limits, data integrity, and security. AI lowers the barrier to getting a working first draft, but the judgment to know whether that draft is safe and correct still requires hands-on knowledge.
For an Admin trying to build logic skills, what's a practical starting exercise beyond reading documentation?
Taking a broken or over-engineered flow in a sandbox org and tracing exactly why it fails — or why it works — builds the diagnostic instinct faster than building from scratch. Asking an AI to explain each decision point in an existing automation, then trying to predict what would happen if a condition were removed, trains systematic thinking without the pressure of a live build. The goal is developing a mental model of cause and effect, not memorizing syntax.
Why is the BA starting point described as undervalued compared to Admin or Dev when BAs typically have no Salesforce-specific credentials?
Stakeholder management, requirements translation, and political navigation inside organizations are skills that take years to develop and are genuinely difficult to teach — whereas Salesforce platform mechanics can be learned in months, especially with AI assistance. Most technical Salesforce professionals have the inverse problem: they can build but struggle to surface the right requirements or manage competing priorities from business users. The credential gap is real but closeable; the business acumen gap is the one that quietly derails implementations.