Nobody’s pricing in the cost of staying still. Here’s the conversation around Salesforce forms that made me realize how invisible that cost actually is.
I had a conversation this week that I haven’t been able to stop thinking about.
One of our account executives was working through a decision with a customer in higher education. The question on the table was whether to keep their existing form tool – Form Assembly – or move to a native Salesforce alternative. As it happens, the native alternative in the conversation was Revecast Forms, which we built. We have a financial interest in you choosing native over external. Read everything that follows with that in mind. If the argument only holds up when you forget who’s making it, it’s not a good argument.
With that on the table – the customer’s reasoning for staying with Form Assembly was completely sound. It’s cheap. It costs them only about a thousand dollars a year, it rarely breaks, and as they put it, it’s been “one of our least problematic” tools. Evaluating and approving a new product is a long process. They’re happy with what they have.
Every one of those statements is true. And I want to be clear about that, because the easy version of this post is the one where I tell you legacy tools are bad, our tool is good, and the people who don’t switch are being short-sighted. That version would be dishonest, and you’d see through it.
The customer staying on Form Assembly isn’t being short-sighted. They’re optimizing for the things they can see – cost, stability, enablement – and legacy wins on all three. The problem is the thing they can’t see. And the reason they can’t see it is because it’s a capability that doesn’t exist yet.
That’s what I want to walk through. Not because I think one customer made one wrong call, and not because I’m trying to sell you Revecast Forms. Because I’ve now had some version of this conversation enough times that I think it’s a sector-wide blind spot, and I think it’s about to age very badly.
What an External Form Actually Costs Your Salesforce Strategy
Here’s the technical reality: Form Assembly, and other tools like it, are not native to Salesforce. The form doesn’t live inside Salesforce. Instead, it lives outside it and connects in. In practice, when a form like this is embedded into an application or recruitment process, it runs as an external application sitting inside your page. What matters is where the form’s logic and field-level state actually live: in the external tool, not on the Salesforce platform. The data syncs back to Salesforce, but the sync isn’t real time. There’s a handoff. A boundary. A piece of the experience that Salesforce can see the results of, but can’t see into.
For years, that boundary didn’t matter. A form was a form. The applicant filled it out, the data landed in your CRM, everyone moved on. If the sync happened in real time or thirty seconds later, who cares.
Here’s what changed. The boundary that didn’t matter for data sync turns out to matter enormously for intelligent agents.
An Agentforce agent – the kind of AI agent we build for clients every day – can only help with what it can see. If the application form is native to the platform, the agent knows what step the applicant is on. It knows which field they’re stuck on. It can answer a question about that specific field, right now, in context. It can guide them through. It can notice they’ve abandoned the form halfway and follow up. In our higher education use case, it can give a prospective student a genuinely connected experience.
If the form runs in an external system – even when it’s rendered inline on your own page – the agent is standing on the outside of the glass. The form looks like it’s part of your page, but its live state isn’t. The agent can’t see what step the applicant is on. It can’t read the field they’re stuck on. It can’t help them fill it out, because from the agent’s perspective, the form is a black box bolted onto the page.
The capability isn’t degraded with a non-native form. It’s not “slower” or “a little worse.” It’s architecturally impossible.
You cannot put an intelligent agent outside a window it cannot see through. No amount of configuration fixes it. The limitation is the boundary itself, not the brand on the boundary. Any external, connector-based form has this problem. The reason native tools like Revecast Forms don’t is because they live on the platform, so there’s no glass for the agent to stand behind.
Your Salesforce Form Tool Has an Agentforce Problem
So here’s the trap. When that customer evaluated Form Assembly against the native option, they compared the things both tools do today. Cost. Stability. Enablement. Form Assembly won, fairly, on the criteria that were visible.
What wasn’t on the comparison sheet was the agent-guided application experience – because at the moment of the decision, nobody was building that. You can’t weigh a capability you don’t know to ask for. The cost of choosing an off-platform form isn’t a line item that shows up next year. It’s a door that quietly closes, and you don’t hear it close, because you weren’t planning to walk through it yet.
Twelve to eighteen months from now, when this same institution wants an intelligent agent guiding students through admissions – and they will, because everyone will – they’ll discover the form they chose for a thousand dollars a year is the exact thing standing between them and the experience they want to deliver. At that point, re-engineering it isn’t a thousand-dollar problem anymore.
The decision that looked like “save money, keep what works” was actually “trade a future capability we couldn’t see for a present cost we could.” Nobody framed it that way. Nobody could have, because the future capability hadn’t arrived. That’s what makes it a blind spot rather than a mistake.
This Isn’t One Organization. It’s the Whole Sector.
I want to be careful not to make this about one customer, because the thing that actually alarms me is that it isn’t.
I’ve interviewed hundreds of architects from other firms working in the education and non-profit ecosystem. I ask them what they’re building applications and student-facing processes with. The answers are remarkably consistent. They’re on Formstack, JotForm, Form Assembly – the same generation of tools, the same architectural pattern. A form that lives somewhere and connects in.
Every one of those choices was reasonable when it was made. These are mature, stable, well-understood tools, and the people who chose them are good architects making defensible calls. The entire sector defaulted to a pattern that was completely fine for a world without intelligent agents, but that world is ending. And because everyone made the same call for the same good reasons, the whole ecosystem is going to hit the same wall at roughly the same time.
The students are the ones who lose in the meantime. Right now, a prospective student filling out an application is largely on their own. They get stuck on a field, they guess, or they give up, or they email an office that replies in three days. The technology to put a patient, knowledgeable guide inside that moment – answering the question they have about the field they’re looking at – exists today. It’s just locked out by an architectural decision made years ago for reasons that had nothing to do with students.
The Part Where I Argue Against Myself
Here’s where I’m supposed to tell you to rip out Form Assembly and buy Revecast Forms. I’m not going to do that.
Sometimes staying is the right call. If an institution has no near-term intention of deploying student-facing agents, and the form works, and the budget is tight, then a thousand dollars a year for a tool that rarely breaks is a perfectly good answer. Migration is real work. Education Cloud’s data model is genuinely more complex, and rebuilding Salesforce forms to be native isn’t free. The switching cost is not imaginary, and anyone who tells you it is – including me – is selling you something.
I’ll go further. Native isn’t automatically the answer to everything, and Revecast Forms isn’t the right call for every org. There are integrations where an external tool genuinely beats the native option, and a good architect should fight for those. The point of this piece is not “native good, external bad,” and it’s definitely not “our product good.”
The point is narrower and, I think, more useful: the criteria you use to evaluate a tool today needs to include the capabilities you’ll want tomorrow, even when those capabilities aren’t on anyone’s roadmap yet. The mistake isn’t choosing Form Assembly. The mistake is choosing it without anyone in the room asking, “What does this make impossible later?” – because that question is the one that the cheap, stable answer is specifically designed not to raise.
What I’d Actually Ask You to Do
If you’re an architect or an administrator making one of these decisions, I’m not asking you to switch tools, and I’m not asking you to call us. I’m asking you to add one question to the evaluation: if we want an intelligent agent to guide people through this process in two years, does this choice make that possible, hard, or impossible? If the honest answer is “impossible,” that belongs on the comparison sheet next to the price.
If you’re an account executive and you’re watching a customer choose the cheap, stable, comfortable option, I’m not asking you to scare them into a migration they don’t need. I’m asking you to make sure the door they’re closing is a door they know they’re closing. The decision is theirs. The visibility into what it costs is your job.
And if you’re a prospective student somewhere, filling out an application right now, getting stuck on a field with no one to ask – you’re the reason any of this matters. The boundary that’s invisible to the people choosing the tools is the wall between you and the help you should be able to get. That’s the whole point of building these systems.
The form is the cheapest thing on the invoice. It might be the most expensive decision on the page.
Learn more about Revecast Forms.
