strategy-and-roadmap

Compare original and translation side by side

🇺🇸

Original

English

🇨🇳

Translation

Chinese

Strategy and roadmap (your plan)

策略与路线图（你的行动计划）

A pile of tactics is not a go-to-market. This turns your brief into a diagnosis and a prioritized plan, and becomes the one place you come back to when you don't know what to do next.

Use this when: you've done the brief and don't want to guess which skill to pull, or a chat ended and you're staring at "what now," or you keep getting handed standalone tasks (do the homepage, make a content plan) that don't add up to a strategy.

一堆零散的策略构不成上市（GTM）方案。本流程将你的简报转化为诊断报告和优先级明确的计划，当你不知道下一步该做什么时，这里就是你的唯一参考。

适用场景： 你已经完成了简报，但不确定该选用哪个skill；或者对话结束后你陷入“接下来该做什么”的迷茫；又或者你不断接到孤立的任务（比如做首页、制定内容计划），但这些任务无法形成完整策略。

The core idea

核心理念

The other skills are moves. This is the game plan that decides which move, and when. It does three things:

Diagnose. Read the brief honestly. Where are you actually strong? What is the single biggest thing standing between you and your next 10 users or your first dollar? Name one bottleneck, not five.
Prioritize by stage. The right first move for a founder with zero users is not the right first move for one with 500 users and no revenue. Sequence the work to the stage you're actually at, and say out loud what to ignore for now.
Persist. Write it to
```
docs/gtm-cofounder/gtm-roadmap.md
```
so it survives the chat. Next session you read the roadmap, not a dead thread.

其他skill都是具体行动，而本流程是决定行动顺序和时机的全局计划。它主要完成三件事：

诊断：坦诚解读简报。你真正的优势在哪里？阻碍你获得下10个用户或第一笔收入的最大瓶颈是什么？只指出一个核心瓶颈，而非五个。
按阶段排序：对于零用户的创始人来说，正确的第一步行动，和拥有500用户但无收入的创始人的第一步行动截然不同。根据你当前所处的阶段安排工作顺序，并明确说明现阶段可以忽略的事项。
持久化：将计划写入
```
docs/gtm-cofounder/gtm-roadmap.md
```
，使其在对话结束后依然保留。下次会话时，你只需查看这份路线图，而非已结束的对话线程。

How to run this (for the agent)

执行方式（面向Agent）

Read
docs/gtm-cofounder/founder-brief.md
first. If it does not exist, run
```
start-here
```
before this. Everything here keys off the brief.
Lead with a diagnosis, not a task list. Two or three sentences: here is what's working, here is the one bottleneck that matters most right now, and here is why.
Be willing to tell them to do almost nothing. If most of the brief is
```
[assumption]
```
and they've barely spoken to users, the honest roadmap is short: go talk to 10 users, come back. Do not pad it to look thorough.
Sequence into three horizons: Now (the one thing), Next (the two or three that follow), Later (parked on purpose, so it's off their mind). Map each item to a skill.
Offer, never impose. Present the roadmap and ask what they want to start with. Do not assign or begin a task they have not chosen. If they say no to something, move it to Later and drop it, do not resurface it next turn.
Write the plan, not a copy of the brief.
```
docs/gtm-cofounder/gtm-roadmap.md
```
is a separate, different document from
```
docs/gtm-cofounder/founder-brief.md
```
. The brief is the context; the roadmap is the short, decision-shaped plan you derive from it. Never re-save the brief's contents into the roadmap. If
```
docs/gtm-cofounder/gtm-roadmap.md
```
does not open with a one-line Diagnosis followed by Now / Next / Later, it is wrong.

先阅读
docs/gtm-cofounder/founder-brief.md
：如果该文件不存在，请先执行
```
start-here
```
流程。本流程的所有内容都基于这份简报。
先给出诊断，而非任务列表：用两到三句话说明：哪些方面运作良好，当前最关键的瓶颈是什么，以及原因。
敢于建议几乎不做任何额外事：如果简报中大部分内容都是
```
[assumption]
```
（假设），且创始人几乎没有与用户交流过，那么坦诚的路线图应该很简短：去和10位用户交流，之后再回来。不要为了显得全面而添加无关内容。
分为三个阶段排序：Now（当下唯一要做的事）、Next（接下来的两到三件事）、Later（故意延后的事项，让创始人不必分心）。为每个事项匹配对应的skill。
提供选择，而非强制安排：展示路线图并询问创始人想从哪项开始。不要分配或启动创始人未选择的任务。如果创始人拒绝某件事，将其移至Later并搁置，下次会话不要再提及。
撰写计划，而非复制简报内容：
```
docs/gtm-cofounder/gtm-roadmap.md
```
是与
```
docs/gtm-cofounder/founder-brief.md
```
独立且不同的文档。简报是背景信息，路线图是从中提炼出的简短、决策导向的计划。切勿将简报内容重新保存到路线图中。如果
```
docs/gtm-cofounder/gtm-roadmap.md
```
不是以一行Diagnosis（诊断）开头，接着是Now / Next / Later，那么这份文档就是错误的。

Stage decides the first move

阶段决定第一步行动

Use the brief's stage and evidence answers to pick the starting point. A rough guide, not a law:

Where they are	The bottleneck is usually	Start with
Pre-users, mostly `[assumption]`	You don't yet know who it's for or whether the pain is real	`talk-to-users` , then `who-is-this-for`
A few users, can't describe them	Fuzzy ICP, so nothing else can be sharp	`who-is-this-for` , then `positioning-and-story`
Users but the message is generic	Positioning and homepage speak to no one	`positioning-and-story` , `value-prop-that-converts` , `the-homepage`
Good product, nobody arrives	Distribution: no first channel that works	`first-50-users` , then `launch-it`
Developers love it, nobody pays	The buyer vs user gap	`market-to-devs-sell-to-buyers`
Growing but flying blind	No read on what's actually working	`know-if-its-working`

When in doubt, the default first move for an early founder is almost always

talk-to-users

. Real evidence beats a clever plan.

根据简报中的阶段和实证回答选择起点。以下是大致指南，而非硬性规则：

所处阶段	通常的瓶颈	从以下步骤开始
无用户，内容多为 `[assumption]`	尚未明确目标用户或痛点是否真实	`talk-to-users` ，然后是 `who-is-this-for`
有少量用户，但无法清晰描述用户画像	ICP（理想客户画像）模糊，导致其他工作无法精准开展	`who-is-this-for` ，然后是 `positioning-and-story`
有用户，但推广信息过于通用	定位和首页内容无法触达特定人群	`positioning-and-story` 、 `value-prop-that-converts` 、 `the-homepage`
产品不错，但无人问津	获客渠道：没有有效的初始渠道	`first-50-users` ，然后是 `launch-it`
开发者喜爱产品，但无人付费	用户与付费者之间的差距	`market-to-devs-sell-to-buyers`
用户增长，但无法掌握实际运营情况	不清楚哪些策略真正有效	`know-if-its-working`

若有疑问，早期创始人的默认第一步行动几乎总是

talk-to-users

。真实的用户反馈胜过精妙的计划。

The roadmap file

路线图文件

Save to

docs/gtm-cofounder/gtm-roadmap.md

in the founder's project (create the

docs/gtm-cofounder/

folder if it does not exist), using

gtm-roadmap.template.md

in this repo. This is a distinct file from the brief and must contain the plan, not the founder's answers. Keep it short and living:

Diagnosis: the one bottleneck, in a sentence.
Now: the single next move, the skill it maps to, and what "done" looks like.
Next: two or three moves queued behind it.
Later: parked on purpose (things like the homepage or a content plan that are real but not yet).
Log: what got done and what was learned, so the plan improves instead of repeating.

将文件保存到创始人项目的

docs/gtm-cofounder/gtm-roadmap.md

（如果

docs/gtm-cofounder/

文件夹不存在则创建），使用本仓库中的

gtm-roadmap.template.md

模板。这是与简报不同的独立文件，必须包含计划内容，而非创始人的回答。保持文件简短且可动态更新：

Diagnosis（诊断）：用一句话描述当前最大的瓶颈。
Now（当下）：唯一的下一步行动、对应的skill，以及“完成”的标准。
Next（下一步）：后续排队的两到三项行动。
Later（延后）：故意搁置的事项（比如首页或内容计划，这些是真实需求但暂不紧急）。
Log（日志）：记录已完成的工作和学到的经验，以便计划不断优化而非重复无效工作。

Coming back to it (every session)

后续回顾（每次会话）

This is the fix for "the chat ended and I don't know what to do next." Start each session here:

Read

docs/gtm-cofounder/founder-brief.md

and

docs/gtm-cofounder/gtm-roadmap.md

Say where things stand and the one next move, in a sentence or two.
Let the founder pick: do the next move, change priorities, or park something.
After the work, update
```
docs/gtm-cofounder/gtm-roadmap.md
```
: move finished items to the log, promote the next one, capture what was learned.

You are not a menu of skills they have to operate. You're the co-founder holding the plan.

这是解决“对话结束后不知道下一步该做什么”的方案。每次会话都从这里开始：

阅读

docs/gtm-cofounder/founder-brief.md

和

docs/gtm-cofounder/gtm-roadmap.md

。

用一两句话说明当前进度和下一步行动。
让创始人选择：执行下一步行动、调整优先级，或搁置某项任务。
完成工作后，更新
```
docs/gtm-cofounder/gtm-roadmap.md
```
：将已完成的事项移至日志，推进Next中的事项，记录学到的经验。

你不是供创始人选择的skill菜单，而是持有全局计划的联合创始人。

Your next 30 minutes

接下来30分钟的任务

Make sure

docs/gtm-cofounder/founder-brief.md

exists. If not, run

start-here

Write a one-sentence diagnosis: the single biggest bottleneck right now.
Fill
```
docs/gtm-cofounder/gtm-roadmap.md
```
: one Now, a short Next, an honest Later.
Pick the Now move and open its skill. Just that one.
Next session, come back here first. Don't start from a blank chat.

Built from real dev-tool GTM experience, with frameworks from Adam Frankl (The Developer-Facing Startup) and Jakub Czakon (markepear.dev). When a framework can't make the call, that's what a human is for: The DevTool GTM Company.

确保
```
docs/gtm-cofounder/founder-brief.md
```
存在。如果不存在，执行
```
start-here
```
流程。
撰写一句话诊断：当前最大的瓶颈。
填写
```
docs/gtm-cofounder/gtm-roadmap.md
```
：一个Now事项，简短的Next列表，坦诚的Later事项。
选择Now事项并打开对应的skill。只做这一件事。
下次会话时，先回到这里。不要从空白对话开始。

基于真实的开发者工具GTM经验构建，融合了Adam Frankl（《面向开发者的创业公司》）和Jakub Czakon（markepear.dev）的框架。当框架无法做出决策时，就需要人类的判断：The DevTool GTM Company。