Most business owners assume a big pivot needs a big budget. It doesn’t. The businesses growing fastest right now are the ones testing new directions for a few hundred dollars and a week of their time, then deciding what to build based on real answers instead of guesswork.
Why treating your business as already dead works
ScoreApp co-founder Daniel Priestley, who has launched multiple companies, uses a deliberately uncomfortable framing with his own team on a regular basis: treat the business as if it’s already dead, then ask what would be built differently starting from scratch today. Would the same people be hired into the same roles? Would the same products, pricing, and ads survive a rebuild? Asking that question surfaces decisions that quietly went stale months ago.
Priestley says, “In the age of AI, everything is moving so fast that your business is already dead. Every decision that you made more than a few months ago is completely outdated.” That’s not pessimism. It’s a prompt to keep reinventing rather than defending choices made under different conditions.
Signs a business needs reinventing
A handful of patterns tend to show up together when a business has drifted out of date. Selling time for money instead of packaging an outcome is one. Relying on slow, traditional marketing while faster channels pull attention elsewhere is another. Selling a process rather than a result usually turns into a race to the bottom, and customers who fight hard on price are often signalling that the offer has become commoditized.
None of these signals are dramatic on their own. A single price objection or one slow-moving campaign doesn’t mean a business is finished. It’s the pattern that matters. When several of these show up together over a few months, that’s usually a sign the underlying model needs a rethink rather than a tweak to the sales script.
- Revenue depends on hours worked rather than results delivered
- Marketing moves slower than the market itself
- The offer describes a process instead of an outcome
- Price is the first thing prospects push back on
None of these are fatal. They’re just signals that it’s time to test something new, and testing doesn’t have to be expensive.
The expensive way to test an idea
Picture two agency owners, each with fifteen years running services and each spotting the same opportunity: software is now cheaper and faster to build than ever, so why not launch a product? The first owner hires a small development team, leans on AI coding tools to keep costs down, and spends two to three months building before trying to sell anything. By the time the first customer signs up at around twenty dollars a month, the team has been paid for roughly three months of work. The math rarely favours the builder in that scenario. A five figure spend to land a single twenty dollar sale is a common outcome of skipping validation.
The fast, cheap version of the same test
The second owner runs the experiment differently. A simple landing page goes up promoting a waiting list, framed loosely: a software product is coming later this year, and early sign ups get a say in what it includes. The page and list cost next to nothing to launch. A few hundred dollars goes toward social ads, a handful of posts go out, and an email lands in the existing customer database.
Within a week, five hundred people have joined and answered five to ten questions about pricing, features, and the problem they’re trying to solve. No product has been built and no development budget has been spent, yet the data now shows what to charge, what to build first, and whether the whole idea deserves further investment. Some waiting lists reveal that the original plan needs a complete rework before a single line of code gets written, which is a far cheaper place to discover that than three months in.
Building that kind of test used to require a developer and a few days of setup. Priestley built ScoreApp’s template library and AI builder around exactly this problem, so anyone can start from a ready-made template and have a similar page collecting real answers within the hour instead of writing custom code.
Four fast, cheap experiments worth running
Priestley points to four formats he returns to again and again because each one produces useful data in days rather than months.
Waiting list
A waiting list keeps the offer deliberately vague while gathering names alongside answers to pricing and feature questions. It works because people join out of curiosity, and the questions asked while they join do the real work.
Online assessment
An assessment asks people to describe the problem they’re trying to solve and then sorts their answers into categories. The product doesn’t need to exist yet. Anyone who completes the assessment has already shown they’re part of a willing market for whatever gets built next.
Discussion group
A pop-up group on WhatsApp or Facebook, sometimes running for as little as ten days, gets people talking directly. Priestley has seen discussion groups grow to between five and ten thousand members before a single product launched, generating months of insight along the way.
Mini course
Twenty or thirty minutes of recorded video, packaged behind an opt-in wall, tests demand for free while doubling as a lead capture mechanism. It’s cheap to produce and useful for measuring how much it costs to generate a single lead.
Each of these formats overlaps with what a well-run waiting list can achieve when the right questions get asked at sign up, and both approaches lean on the same idea: collect real answers before spending real money.
Turning the data into a consulting-grade report
The value of these experiments compounds once the data gets analysed properly. Priestley describes collecting one hundred responses from a waiting list, exporting the raw data, and feeding it into Claude with a request to summarise what it showed. The output identified four distinct market segments, each with a different core problem, ranked them by priority, and listed typical price points and competitors to benchmark against.
That kind of output used to require hiring a research firm. Getting it from a spreadsheet and an AI prompt changes the economics of validation entirely. It also reflects the same principle Priestley built into ScoreApp itself: answers people give directly, rather than behaviour inferred from clicks, tend to reveal fit and readiness far more reliably than guesswork. ScoreApp’s scoring and segmentation tools are built around exactly that idea, turning raw answers into the kind of clarity a consulting report would normally charge thousands for.
What this looks like for different businesses
An agency with fifteen years of client relationships already understands how its customers think and where they spend time online. That’s a strong starting point for spinning out a software product tested through a waiting list rather than a costly build. Agencies already exploring this route can see practical ways to turn client relationships into new revenue without adding headcount or overhead.
A coach or consultant might test a group program or an annual retreat the same way, gauging demand before committing to logistics. A software company could test a bootcamp or accelerator that sits alongside its existing product, using a short assessment to see who would actually sign up. None of these tests require touching the core business. They run alongside it, quietly, until the data says otherwise.
The common thread across all three is speed of feedback. Businesses ready to test a similar idea can try ScoreApp free and have a working experiment live within the hour rather than spending weeks on a build that might not land. Pricing plans that fit a small experiment through to a full campaign are laid out on the ScoreApp pricing page, so the cost of testing stays predictable from the start.
Start with one experiment this week
Pick one idea sitting on the back burner, choose the cheapest format that could test it, and set a deadline of seven days to get real answers. A waiting list, an assessment, or a short course will tell more about demand than months of internal debate ever could.
