Validating the market before spending months on development
SaaS / AI Tool
Pre-development startup with a product hypothesis: an AI tool to automate a repetitive task for marketing and communications teams. The skills to build it were there. What was missing was knowing whether the market wanted it, why, and how much it would pay for it.
Before writing a line of code we tested three different promises, time savings, fewer errors and process scalability, each with its own landing page and campaigns aimed at professional audiences. Anyone joining the waitlist also filled in a micro-survey on use case, how often the problem shows up, tools already in use and monthly spend capacity.
1,240 signups at €1.82 each, 18% high intent. The answers surfaced a usage segment the team hadn't considered, 34% of the most interested users. And the time-savings promise performed 40% better than the other two.
1,240
Waitlist signups
€1.82
Average cost per signup
18%
High-intent users
3
Value propositions tested
The context
Pre-development startup. The idea was reasonably clear: an AI tool to automate a repetitive part of the work marketing and communications teams do.
All the skills to build it were there. What was missing was the rest: whether the market actually wanted it, why it wanted it, and how much it would pay for it.
The problem
Building first and validating later is the most expensive way to be wrong, and in software it's also the most common. The temptation was to start with development in order to have something to show, at the risk of reaching six months of work on a product the market doesn't want, or doesn't want in that form, or at that price.
What we built
We tested three promises instead of one. Time savings, fewer errors, process scalability: three landing pages with different headlines, tone and lead benefit, with Meta traffic split evenly to see which angle landed best.
Joining the waitlist didn't ask for money, but it did ask for something. A micro-survey on main use case, how often the problem shows up, tools already in use and monthly spend capacity. That survey is what separates curiosity from real commercial interest: an email address on its own doesn't tell you.
The campaigns were optimized for form completion rather than clicks, on specific professional audiences, marketing managers, content creators, agencies. Not "people interested in artificial intelligence".
The result
1,240 signups at an average of €1.82, with 18% classified as high intent: 223 people who reported a real problem, a workable budget and concrete urgency.
The useful part, though, came from the survey answers. A usage segment surfaced that the team hadn't considered, worth 34% of high-intent users. The time-savings promise performed 40% better than the other two.
Positioning, priorities and roadmap were revised with that in hand, before development. The value of this project isn't the waitlist: it's the six months of work nobody spent going the wrong way.