Service
App Prototyping
From idea to clickable prototype in days.
Before six-figure budgets go into a product, you should know whether anyone will use it. In one to two weeks I build a real, clickable prototype. Good enough for user tests, investor conversations, and the internal go/no-go call.
- A clickable prototype, not a slide deck
- Core flow usable end to end
- User tests and analysis
- A clear recommendation: build or don't
Before the budget flows
The most expensive software projects are the ones built without anyone checking whether the result gets used. A concept on slides doesn't answer that question, because nobody can operate a slide.
A clickable prototype does. Not as a slideshow, but as something you can hand to someone and ask them to complete a task. What happens next teaches you more than any discussion round.
What products actually fail on
CB Insights analysed 431 failed VC-backed companies that shut down from 2023 onwards. The most common reason is "ran out of capital" at 70 percent, but that's the cause of death, not the illness.
The actual reason sits right behind it: 43 percent failed on poor product-market fit. They built something nobody wanted urgently enough in that form. The money ran out because the product didn't pull.
That question can be answered beforehand, and far more cheaply than afterwards. Not through market research on slides, but by putting something in the hands of people from the target group and watching what happens.
- Ran out of capital
- 70%
- Product-market fit
- 43%
- Bad timing
- 29%
- Unit economics
- 19%
Multiple reasons per company are possible, so the values add up to more than 100 percent. The highlighted row is the one a prototype can test in advance.
Source: CB Insights, 431 VC-backed shutdowns since 2023, March 2026
How much testing is enough
A common worry is that user testing is elaborate and needs many participants. For the kind of insight at stake here, that isn't true.
The Nielsen Norman Group recommends five people from the target group for qualitative testing. After that the observations repeat, and the value of each further participant drops steeply. Their recommendation: three small rounds of five rather than one large round of fifteen, because you can change something between rounds.
The distinction matters: five people surface usability problems. For statements about demand, willingness to pay, or market size you need considerably larger samples and different methods.
What you get
Within one to two weeks, a usable prototype of the core flow, good enough for user tests, investor conversations, and the internal go/no-go decision. Plus an analysis of the tests and a clear recommendation.
That recommendation can be: don't build it. That isn't a wasted project, it's the cheapest way to avoid an expensive one.
Frequently asked questions about this service
- How is this different from a Figma clickdummy?
- A clickdummy shows predrawn screens in a fixed order. A real prototype responds to input, keeps state, and allows detours. The difference shows immediately in user testing, because people rarely take the intended path.
- Can the prototype be reused later?
- Partly. Interface and flow logic can usually be carried over, the data layer rarely, because a prototype deliberately takes shortcuts. I say upfront what is throwaway work and what stays, so you can plan for it.
- How many users make a meaningful test?
- For fundamental usability problems, five to eight people from the target group suffice; after that the observations repeat. For statements about demand or willingness to pay, you need considerably more and different methods.
Other services
Have a similar project in mind?
Tell me about it. I'll reply within 24 hours with an honest assessment.