Pretotyping

There’s a well-known saying in product: build the right thing, then build it right. Pretotyping is about the first half - and most teams skip it entirely. The concept comes from Alberto Savoia, a former engineering director at Google, who noticed a pattern in product failures: most of them weren’t execution failures. The team shipped on time, the code was solid, the design was clean. The product just failed because nobody wanted it 😬 He called this the “it” problem - you built the thing right, but you didn’t build the right it. He wrote it all up in The Right It, which is worth reading before you start your next initiative.

​

Pretotyping vs. prototyping

The names are close enough to cause confusion, but the goals are completely different. A prototype tests whether your solution works - whether the UX makes sense, whether the technology is feasible, whether people can complete the flow. It assumes the problem is real and asks “can we solve it?” A pretotype tests whether the problem is real in the first place. It comes before the prototype. It asks “should we even bother?” You’re not testing the solution - you’re testing the assumption that enough people have this problem badly enough to want a solution at all. Prototypes are relatively expensive to build. Pretotypes are intentionally cheap. The whole point is to get signal with minimum effort before you commit resources.

​

The law of market failure

Savoia’s central observation is uncomfortable but important: the vast majority of new products fail, even when built by smart people with good intentions and adequate resources. The reason is almost always the same - insufficient evidence that the market wanted the thing. Pretotyping is the antidote. Instead of trusting your intuition or your stakeholders’ enthusiasm, you go find real data from real people before you build.

​

Pretotyping techniques

There are several ways to pretotype, ranging from almost zero effort to a few days of work. The right choice depends on what assumption you’re testing. The fake door - Put a button, link, or sign-up form for a feature that doesn’t exist yet. Measure how many people click it. If nobody does, that’s important signal. If everyone does, you have a waiting list before you’ve written a line of code. Works especially well if you already have an existing product or audience. The landing page - Build a simple page describing the product as if it already exists. Include a call to action - sign up, join the waitlist, even a buy button. Drive some traffic to it and measure conversion. The gap between “sounds interesting” and “I’ll give you my email” tells you a lot. The concierge - Do manually what your product would eventually do automatically. If you’re building a service that matches freelancers to projects, do the matching yourself first. You learn what customers actually need, you generate real revenue signal, and you don’t spend six months building a platform before you know if anyone wants it. The impersonator - Use an existing off-the-shelf tool to simulate your product before building your own. If your idea is a specialised CRM, test it with a customised Airtable first. If it’s a content curation newsletter, test it manually with a Substack. You’re not building to scale - you’re building to learn. The pitch - Tell people about the product and ask them to commit in some way - pre-pay, sign a letter of intent, refer a friend. Words are cheap but money and commitment are real signal. If nobody will pay for it before it exists, think carefully about whether they’ll pay for it after.

​

What you’re looking for

A pretotype gives you one of three outcomes: Green light - Strong positive signal. People click, sign up, pay, refer friends without being asked. You have evidence of demand. Move forward with more confidence. Red light - Weak or no signal. Nobody bites. The market is telling you something. Either the problem isn’t real, the solution isn’t compelling, or you’re reaching the wrong people. Don’t ignore this. Yellow light - Mixed signal. Some interest but not overwhelming. This usually means you need to sharpen something - your positioning, your target audience, your value proposition - before you can get a clearer read. Lesson learned: the temptation when you get a red or yellow light is to explain it away. “The landing page wasn’t good enough.” “We didn’t drive enough traffic.” “The timing was off.” Sometimes that’s true. More often it’s the market being honest with you and you not wanting to hear it.

​

The pretotype mindset

The hardest thing about pretotyping isn’t the technique - it’s the mindset. You have to genuinely want to find out you’re wrong, not just go through the motions of validation while hoping for confirmation. Savoia’s framing is helpful here: your job is to collect “your own data” (YOD) - real behavioural evidence from real people, not opinions, surveys, or gut feel. People lie in surveys. People don’t lie with their wallets or their time. If you’re starting something new, run at least one pretotype before you write a line of code. It’s the cheapest insurance policy in product 🛡️