做产品 PMaker
空格的键盘
他现在的做法
日期客户金额备注
3/14李1240未收
3/16王388已收
3/1?张2715忘了
手工维护的表格 每周花两小时
这张表
就是需求文档
它告诉你的
· 他要的字段只有这四个
· 状态只有三种,不用做流程引擎
· 痛点是「忘了」,不是「录入慢」
· 每周两小时,是你的价值上限

用户凑合出来的那个东西,是一份已经被真实使用验证过的需求说明书,而且是免费的。

Find the Existing Workaround First

If a problem really exists, the user is already solving it with some clumsy method. Finding that clumsy method is worth more than ten interviews.

What you will run into
  • 做出来的功能比用户实际需要的复杂五倍
  • 用户说要,但你说不清他现在没有这个东西是怎么活的
  • 上线后用户还是回去用他原来那套

No workaround means one of two things: the problem doesn't hurt, or it hurts so much no one can get by. The first, skip. The second is a big opportunity. The way to tell them apart is whether they've paid any other price for it — say, just stopped doing the thing at all, or paid someone to do it.

Conversely, if the workaround is already good enough, don't build it either. For users to switch from something that works to you, the payoff usually has to be several times larger — not just a little better.

Where to Look

找什么它意味着什么
自己搭的表格需求最清楚的一种。列名就是字段,颜色就是状态
一堆截图和微信收藏他在做的是「攒起来以后找得到」,核心诉求是检索
写了个小脚本能写脚本说明痛到愿意学。这类用户是最好的早期用户
雇了个人在做已经在付钱了,且能算出具体金额。定价的锚点
几个工具串着用你的机会在那几个工具的接缝处,不在其中任何一个里面
干脆放弃了这件事需求存在但成本太高。做出来可能是全新的市场
The first one is the most easily missed. A column that's missing from a user's spreadsheet is almost always one they genuinely don't need — AI-generated interfaces tend to fill in a pile of fields on their own, and those fields are all dead weight.

How to Read It

Once you have the workaround, break it down along these four lines and the outline of the product falls out.

他保留了哪些信息 表格里有的列 = 必须有的字段 没有的列 = 你别自作主张加上 这一条直接给出数据模型 哪一步最烦 他抱怨的那一步,不一定是耗时最长的 是最容易出错、最怕出错的那一步 这一条给出主任务 他为此付出了多少 每周两小时,或者每月八百块 这个数是你定价的天花板 这一条给出商业模式 他为什么没换掉它 市面上没有,还是有但不合适 不合适在哪,那就是你的切入点 这一条给出差异化
The first one is the most easily missed. A column that's missing from a user's spreadsheet is almost always one they genuinely don't need — AI-generated interfaces tend to fill in a pile of fields on their own, and those fields are all dead weight.