故事 · 一个问句，撬动一个产品
Story · One Question, One Product
斯坦福 d.school 给团队出了个难题：帮发展中国家的早产儿活下来。最直接的问法是 ——「我们如何能造一台更便宜的保温箱？」
这个 HMW 看似合理，却悄悄把答案锁死成了「保温箱」。团队后来重写了问题：
「我们如何能让远离医院的母亲，也能给早产儿保暖？」
就这一句更聚焦于「保暖」而非「保温箱」的问法，把方案空间彻底打开 ——
最终他们做出的不是机器，而是一条可重复加热、像睡袋一样的保暖襁褓(Embrace)，成本只有保温箱的百分之一，救活了数十万婴儿。
这就是 HMW 的魔力：问题问对了，好答案才有空间长出来。问题太窄（保温箱），你永远想不到睡袋。
Stanford's d.school handed a team a hard brief: keep premature babies alive in the developing world. The obvious framing was — "How might we build a cheaper incubator?"
It sounds reasonable, but quietly locks the answer in as "incubator". The team rewrote the question:
"How might we keep premature babies warm when their mothers are far from any hospital?"
That one rewrite — pivoting from "incubator" to "warmth" — blew the solution space open.
What they ultimately shipped wasn't a machine but the Embrace warmer: a reusable, sleeping-bag-like infant swaddle that re-heats with hot water, costs roughly a hundredth of an incubator, and has kept hundreds of thousands of babies alive.
That is the magic of a HMW: frame the question right and good answers have room to grow. Frame it too narrowly (incubator) and you'll never reach the sleeping bag.
1 POV → HMW 生成器 POV → HMW Generator
拼一句问题2 同一需求，三种问法 One Need, Three Framings
太窄Too narrow
—
已经把方案写进问题，没发散余地。
The solution is already baked into the question — zero room to diverge.
刚好Just right
—
聚焦真需求，又给方案留足空间。
Focused on the real need while leaving the solution space wide open.
太宽Too wide
—
大到没有抓手，团队不知从何下手。
So broad there's nothing to grip — the team has no idea where to start.
3 现实里的 POV 与 HMW POV & HMW in the Real World
问题定义：把一堆零散发现收敛成一句话，是从「发现」走向「方案」的关键转折点 —— 双钻的第一个收口。
Problem framing: converging scattered findings into one sentence is the pivot from "discovery" to "solution" — the first closing point of the Double Diamond.
POV 陈述：用户·需求·洞察三段式，逼你说清「为谁、要什么、为什么」，避免空泛的目标。
POV statement: the user-need-insight triplet forces you to spell out who, what, and why — no more vague goals.
HMW 提问：把陈述翻成「我们如何能…」的问句，邀请发散；一个 POV 可拆成好几个 HMW 切入。
HMW question: translate the statement into a "How might we…" prompt that invites divergence — one POV can fan out into several HMWs.
聚焦但开放：好 HMW 既锁定真需求（聚焦），又不指定具体方案（开放）。这是它最难也最值钱的平衡。
Focused yet open: a good HMW locks the real need (focus) without naming the solution (openness). That balance is the hardest and most valuable part.
一句话In One Line
很多团队跳过问题定义，拿着一句模糊的「让产品更好用」就冲去想方案 ——
结果每个人脑子里的「问题」都不一样，方案自然南辕北辙。
POV 和 HMW 的价值，是把问题钉成全队共识的一句话，再用问句的形式邀请所有人往同一个方向发散。
而最深的一层是那个「宽窄」的拿捏：
问题里只要藏了方案（造保温箱），你就再也想不到别的（睡袋）；问题大到没有边界（让世界更好），团队又会原地打转。
好问题落在中间那道窄窄的甜区 —— 聚焦到真需求，开放给所有可能。
Too many teams skip problem framing, grab a vague brief like "make the product better", and sprint straight into solutions —
and then everyone's mental version of "the problem" is different, so the proposed solutions pull in opposite directions.
What POV and HMW buy you is a single sentence that becomes the team's shared problem, then a question that invites everyone to diverge in the same direction.
The deepest layer is the calibration of width:
the moment a solution sneaks into the question (build an incubator), you can never reach the alternative (sleeping bag); blow the question up to "make the world better" and the team just spins in place.
A good question sits in that narrow sweet spot in the middle — focused on the real need, open to every possible answer.
常见误用Common Mistakes
POV 写成功能需求（如「需要一个按钮」）。需求要写人的目标，不是产品的零件。
Writing the POV as a feature spec (e.g. "needs a button"). The need should describe the person's goal, not a piece of the product.
HMW 里藏了方案（如「如何做一个 App」）。问需求别问方案，给发散留空间。
Sneaking the solution into the HMW (e.g. "how might we build an app"). Ask about the need, not the artifact — leave room to diverge.
HMW 大到没法下手（如「如何让人更幸福」）。收窄到具体用户与场景，问题才有抓手。
A HMW so broad it can't be acted on (e.g. "how might we make people happier"). Narrow to a specific user and context — only then does the question give you a handle.