Alberto Savoia explains how to test business ideas fast and cheap before building them. He uses the 'second day sushi' hypothesis as an example of pretotyping, a method to validate whether people actually want what you're planning to make.
People talk a lot about failing fast, but as you can tell from how fast I speak, that's not fast enough for me. Right. I tell them I want you to fail Ferrari fast and Fiat cheap. Sorry, Fiat. Right. So because remember, most new ideas will fail in the market, which means that you have to test a lot of ideas. And if you take six months to test some ideas, good luck, right? Unless your luck is going to take forever. So pretotyping allows you to test very quickly. More importantly, you're not going to experience painful failure. Right? Because you spend $20 to do a prototype test. It's an experiment, right? It doesn't hurt. Definitely doesn't hurt after as much as spending three years and $25 million to build a product that people do not want. So strategy number six, say it with numbers, entrepreneurs, and probably most of you, when you have an idea, you come to me and you express it very vaguely. So here's an idea from some of your fellow students. Maybe five, six years ago, Second day Sushi here's the idea. Packaged sushi is kind of expensive. So they thought, you know what, if we buy sushi that's about to expire, it's only good for an extra eight hours before it kills you. We can buy for 25 cents on the dollars and sell it at 50 cents on the dollars. And since students are young and have a strong stomach, we can handle it. So I said, okay, I'm not going to give my opinion on your idea, but I'm asking them. So this is how they articulated, right people, lots of them will buy not super fresh sushi if it's cheap enough. That's pretty much how they express. I said, look, who are these people? How many is lots and what is cheap enough? So I was in a room actually outside there, just outside this auditorium, and somebody had left a formulas calculus class or electrical engineering class on the wall. And I thought, okay, tell you what, write it like this. X percent of Y will Z. I call this the XYZ hypothesis, right? So it forces you to write your idea in numbers. So in this case, they Translated into number. 20% of pakka sushi buyers will buy secondary sushi if it's half the price of fresher sushi. How do you know if those numbers are right? You don't. It's a hypothesis, right? But at least it articulates and puts into number what your ideas are. And what is the job of a hypothesis. A hypothesis exists to be tested and prototyping our tool to test hypothesis. Last point. Think global Test local. Maybe the second day sushi team is planning to take over the market. Right? Every supermarket is going to have second day sushi. But you need to start to test your ideas very quickly. You want to minimize these metrics that I explained in my book called time to data, dollars to data, distance to data. If people say, I have this great idea and I need six months, $2 million, and I need to fly to Hawaii to do my research, I said, well, no, you're going to do it here. You're going to do it for $20, and I want the results by tomorrow. Right. And a technique for doing that is called hypo zooming. So you take this big hypothesis, 20% of packaged sushi buyers, and you zoom in, you know those documentaries where you see or those videos where you see the earth from space, and then it zooms into a, you know, to a town and then to a particular place and then to a building. I want you to do that in your mind, right? So you take the big XYZ hypothesis and hypo zoom to something you can test here and now. And when I say here and now, I mean it literally here and now. Right. So where are we now? North America. Right. Where in North America? Silicon Valley. Where in Silicon Valley? Stanford University. Where in Stanford University? Building Y2E2. And then ask you, is there a place in this building or nearby that sells sushi? Yes, there is. Upstairs, right? Koopa Cafe. So I said, great, so you're going to hypo zoom. You have your market. A small sample of your market is there. So you can go from the big XYZ hypothesis to the small XYZ hypothesis. 20% of students buying sushi at Kupa Cafe to their dinner will buy second day sushi. Right. And do you believe that this is actually testable? Can I do this test? Absolutely. In fact, we did it. We even did a little video with some of my friends there. Where are you guys? Yes, we did it here. We filmed it. We created little labels that said second day sushi half off. And we slapped them on fresh sushi. Right. It's an impersonator. Pretty type. And then we tried to sell them. In fact, if you're there, that's outside the STVP offices. Now, how many people do you think bought our second day sushi? Zero. Right, right. Because the typical thing is, well, I don't want to get sick. So doesn't mean that second day sushi has no chance. It just means that you need to do some more tests. But frankly, your yoda doesn't look very good.
O processo de reconhecer limites, mesmo quando a vida parece exigir sempre mais.
Depois de tantas histórias, aprendizados e pausas necessárias, o @interior_iza está de volta
