How we built a lead-qualification funnel with Claude, Vercel and Stripe

A scored assessment that gives a prospect a straight commercial read on whether their idea is worth building qualifies that prospect at the same time, before any call happens. Built well, it is a real deployed application: it can run on a host like Vercel with its own domain, take payment through Stripe, and return an on-screen result plus a fuller report. The point is the shape. An assessment that delivers genuine value is also the cleanest lead filter you can have, because the people who complete it and act on it are the people worth your time.

So this is an account of building a funnel that helps the prospect first and qualifies them second, directed through Claude rather than handed to a development team. Here is what such a tool is, how it is put together, and the honest state of the approach.

Why an assessment is a qualification tool

The mechanism worth understanding is that a genuinely useful assessment qualifies leads as a by-product of helping them. Most lead capture is a trade: you give the prospect something thin, a checklist or a generic quiz, in exchange for their email, and then a salesperson chases everyone who entered. That wastes time on both sides, because most of the people who grab a free thing are not buyers.

A real assessment flips that. It gives the prospect commercial value, a straight read on whether their idea survives the subjects that actually decide it, and the act of completing it and engaging with the result is itself the qualification. Someone who works through a serious assessment, pays for the fuller report, and acts on what it says has told you far more about their intent than any form fill could. The people worth your time are the ones who take the work seriously, and the assessment surfaces exactly those people without a single qualifying call.

This is why value-first beats capture-first. A quiz that exists only to harvest emails qualifies nobody, because completing it costs nothing and means nothing. An assessment that delivers a real result self-selects, because only people with genuine intent will engage with it properly. The qualification is not a step you bolt on. It is what happens naturally when the tool is actually useful.

The stack, and what building it involved

A tool like this is a real application, not a page builder template. It can run on a host such as Vercel with its own custom domain, take payment through Stripe, and return both an on-screen result and a fuller report. Content can route people into it, so the writing and the funnel are joined: a reader who reaches the end of a relevant post is one click from the assessment that helps them decide what to do next.

Building it means directing Claude through the whole thing: the scoring logic that turns answers into a commercial read, the deployment onto the host, the payment integration, and the output that presents the result clearly. None of that requires hiring a development team. What it requires is clear thinking about what the assessment has to judge and how to score it, then directing the build and testing the result until it behaves correctly. The skill that matters is specifying the logic and judging the output, not writing the code by hand.

That is the part a business owner should take from the stack. The components, a deployment platform, a payment processor, the application logic, used to mean a project and a developer. Directed through Claude, they become something one operator can assemble and shape. The build is real engineering, but the engineering is now mostly direction and judgement rather than manual coding, which is what changes the maths on building a tool like this for yourself.

Honesty is the tone the funnel depends on

There is a reason to be plain about the state of any automation you put in front of people. Describing automation as done when it is still being refined is exactly the overclaim that makes business owners distrust this entire space, and it is better to lose a little shine than join the pile of tools that promise more than they deliver.

That honesty is not just good manners, it is the whole tone the funnel depends on. The assessment's value is that it tells a prospect the truth about their idea, including when the truth is that the idea does not work. A tool built to tell people uncomfortable truths cannot be marketed with comfortable lies about its own state. The candour about what it is matches the candour it gives, and that consistency is what makes people trust the output.

Why we do not lean on numbers

You will notice this account has no conversion rate, no completion percentage, and no revenue figure. That is deliberate. A figure quoted before it is verified to a standard you would stand behind is provisional at best and misleading at worst. The source rule worth holding to is simple: a number gets stated only if it traces to something solid. Where it does not yet, make the point in plain words instead.

This connects back to the honesty the whole funnel runs on. It would be easy to drop in an impressive-sounding conversion figure and let it do the persuading. It would also be exactly the move that makes the rest of the account suspect. The shape of the argument, value-first assessments qualify better than capture-first quizzes, stands on its logic and on a real build, not on a statistic nobody can back.

What a business owner can take from it

If you rely on sales calls or quote requests to qualify prospects, you know the unqualified ones drain your time. The gap this kind of build closes is that a real assessment, one that genuinely helps the person taking it, filters for intent without a call, because the people who engage with it properly are the people worth talking to. It sits between the generic quiz that qualifies nobody and the custom-coded app that used to need a team, and that middle ground is now reachable. A B2B operator we work with, Pulse Technology Hub, has the same problem most professional-services businesses do, which is sales time lost to prospects who were never going to buy, and a value-first qualifier is the cleanest way to close that gap.

Building one well for a specific business, defining what it should assess, scoring it honestly, deploying it, and wiring it to payment and content, is the work, and it is where the result either qualifies the right people or does not. The way to build it is by directing Claude through the logic, the deployment, the payment, and the output, so it does its filtering quietly while you get on with the prospects who clear it.

The trap of the generic quiz builder

It is worth naming the thing this sits against, because most owners reaching for lead capture reach for a quiz builder first. A quiz builder gives you a few multiple-choice questions and an email box, and it feels like a funnel because it has the shape of one. The problem is that it qualifies on nothing. Anyone will click through five easy questions to see a result, so the people who complete it are not selected for intent, they are selected for idle curiosity, which is the opposite of what you want at the top of a sales process.

The deeper issue is that a generic quiz cannot deliver real value, because it does not actually know anything. It maps your answers to one of a handful of pre-written results, none of which is specific enough to change what you do next. The taker knows this. They have done enough of these to recognise a thin one, and a thin result teaches them the tool was a capture device wearing a quiz costume. That impression attaches to you, which is worse than no impression at all.

A real assessment is the opposite on every count. It asks questions that require thought, it scores them against the subjects that genuinely decide the outcome, and it returns a result specific enough to act on, including the unwelcome result that the idea does not work. That depth is what makes it valuable, and the value is what makes the qualification real, because only someone with genuine intent will work through something substantial and pay for the fuller version. The quiz qualifies for curiosity. The assessment qualifies for intent. They look similar from the outside and do opposite jobs, which is exactly why the cheap version is worse than nothing and the real one is worth building.

A qualifier like this is only as good as the result it hands back, and that result has to read in your voice and hold to your standard every time it runs. That is the wider job Ortopylot does. It is a system that captures your brand so every document and output a business produces, the assessment result included, comes out finished and on brand from one short instruction, which is what lets a tool like this scale without sounding generic. If you want a qualifier that helps prospects and protects your standard at the same time, start at ortopylot.com.