故事 · Dropbox 的一段视频
Story · The Dropbox Video
2008 年,Drew Houston 想做云同步,但「开发它」要耗费几个月、还不知道有没有人要。
于是他没写代码,只录了一段 3 分钟的演示视频,假装产品已经能用,发到极客社区。
一夜之间,等候名单从 5 千人涨到 7.5 万人 —— 假设验证完毕:人们确实极度想要它。
这就是经典的解释视频式 MVP。同样,Zappos 创始人想验证「网上有人买鞋吗」,
没有囤货,只是去隔壁鞋店拍照挂网,有人下单他才跑去买了寄出。
MVP 的灵魂从来不是「做个小产品」,而是用最便宜的方式,问出最贵的那个答案。
In 2008, Drew Houston wanted to build cloud sync — but actually shipping it would take months, with no guarantee anyone wanted it.
So he wrote no code. Instead he recorded a 3-minute demo video that pretended the product already worked, and posted it to a geek community.
Overnight, the waiting list jumped from 5,000 to 75,000. Assumption validated: people desperately wanted this.
That's the textbook video MVP. In the same spirit, Zappos founder Nick Swinmurn tested "will people buy shoes online?"
with no inventory: he photographed shoes at a local shop, listed them on the web, and only ran out to buy them when an order came in.
The soul of an MVP was never "build a tiny product" — it's buy the most expensive answer for the cheapest possible price.
1 5 种 MVP:成本 × 学到什么 The 5 MVPs: Cost vs. What You Learn
点卡片看案例2这一类 MVP:怎么用
验证什么假设The assumption it tests
点上方卡片查看。
真实案例Real-world case
点上方卡片查看。
3 现实里的 MVP 思维 MVP Thinking in Practice
验证假设而非造产品:先问「最关键、最没把握的假设是什么」,再选能最便宜验证它的形态。
Test the assumption, not the product: first ask "what's the most critical, most uncertain assumption?", then pick the cheapest format that can answer it.
最小成本验证:能用人工冒充就别写代码,能用视频就别真做 —— 把钱省到验证之后。
Lowest-cost validation: if humans can fake the back end, don't write code; if a video can prove demand, don't ship the product. Spend the money after the bet is validated.
避免过度开发:闷头造三个月才发现没人要,是创业头号死法;MVP 让你提早撞墙、提早转向。
Avoid over-building: head-down for three months only to find nobody wants it is the #1 startup killer. MVPs let you hit the wall early and pivot early.
学到就算赢:MVP 的产出不是产品,是「学到的东西」 —— 哪怕结论是「这个不行」,也省了大钱。
Learning is winning: the output of an MVP isn't a product, it's validated learning. Even a "no, this won't work" answer saves you a fortune.
一句话In One Line
MVP 里的「P」是 Product,最容易把人带偏 —— 大家盯着「做个产品」,却忘了前面的「最小」和背后真正的目的:验证假设。
高手做 MVP 时先问自己:「如果我错了,最便宜的发现方式是什么?」答案常常根本不需要代码 ——
一段视频、一个落地页、几通人工电话,就够了。
记住:你不是在造一个小产品,你是在花最少的钱,买一个最贵的答案:到底有没有人要。
The "P" in MVP — Product — is what trips everyone up. People fixate on "ship a product" and forget the Minimum in front of it and the real purpose behind it: validate the assumption.
Operators who do this well ask themselves: "If I'm wrong, what's the cheapest way to find out?" The answer often involves no code at all —
a video, a landing page, a few phone calls is enough.
Remember: you're not building a tiny product. You're spending the least possible money to buy the single most expensive answer — does anyone actually want this?
常见误用Common Mistakes
把 MVP 当成「砍功能的简陋产品」闷头开发。先想清要验证哪个假设,能不写代码就别写。
Treating an MVP as a "feature-stripped tiny product" and heads-down building it. Nail down which assumption you're testing first — if you can avoid writing code, do.
不定假设也不定成功标准,做完了不知道学到啥。先写下假设和「多少人下单才算成立」,再开工。
Launching with no hypothesis and no success threshold. Write down the assumption and the bar ("X signups means valid") before you start.
验证通过后还死守那个简陋版不迭代。MVP 是起点不是终点,验证后该投入就投入。
Clinging to the scrappy version after the bet is validated. The MVP is a starting line, not a finish line — once it's validated, invest properly.