npx skills add https://github.com/wondelai/skills --skill improve-retention用于设计可靠改变行为产品的框架。基于一个基本事实:行为无关意志力或动机——它是一个具有可预测方程的设计问题。
福格行为模型 = B=MAP。当动机、能力和提示在同一时刻汇聚时,行为就会发生。
HIGH ┃
┃ ★ Behavior happens
┃ (above the Action Line)
┃
Motivation ┃━━━━━━━━━━━━━━━━━━━━━━━ ← Action Line
┃
┃ ✗ Behavior fails
┃ (below the Action Line)
LOW ┃
┗━━━━━━━━━━━━━━━━━━━━━━━━━
HARD EASY
Ability
行动线: 当动机和能力足够时,提示会引发行为。在线以下,任何提示都无效。这条线是弯曲的:高动机可以弥补低能力,高能力可以弥补低动机。可靠的策略是让行为更容易(向右移动),而不是提升动机(向上移动)。
目标:10/10。 在评审或创建产品行为设计时,根据对以下原则的遵循程度,按 0-10 分进行评分。10/10 表示完全符合所有指导原则;较低的分数表示存在需要解决的差距。始终提供当前分数以及达到 10/10 所需的具体改进措施。
核心概念: 动机是行动的能量。它有三个核心动机,每个都有两面:感觉(愉悦/痛苦)、预期(希望/恐惧)和归属感(接纳/排斥)。动机强大但不可靠——它像波浪一样起伏波动。
为何有效: 动机解释了人们为何想要行动,但它是行为设计中最不可靠的要素。它会飙升(新年决心、产品发布)也会崩溃(第 3 天、第 2 周)。依赖高动机的产品在浪潮退去时就会失败。最好的设计即使在动机处于低谷时也能发挥作用。
关键见解:
广告位招租
在这里展示您的产品或服务
触达数万 AI 开发者,精准高效
产品应用:
| 场景 | 应用 | 示例 |
|---|---|---|
| 新用户引导 | 不要依赖新用户动机峰值能持久 | 设计即使兴奋感消退后也能起效的首次行动 |
| 重新参与 | 假设回归用户的动机较低 | 在要求付出努力之前展示即时价值 |
| 功能采用 | 将动机与难度相匹配 | 简单的功能需要低动机;复杂的功能需要动机助推器 |
| 信息传达 | 利用正确的动机 | 社交健身 → 归属感;金融工具 → 预期(希望) |
| 防止流失 | 诊断是动机下降还是从未高过 | 调查流失用户以了解动机错配情况 |
文案模式:
伦理边界: 切勿制造虚假希望或利用恐惧来夸大动机。动机策略应将用户与真实结果联系起来,而不是制造驱动强迫性使用的焦虑。
深入探讨三个动机、动机波浪和为低谷设计的详情,请参阅:references/motivation-waves.md。
核心概念: 能力是执行行为的能力。它是六个因素中最稀缺资源的一个函数——能力链。如果任何一个环节太弱(太昂贵、太耗时、太令人困惑),行为就会中断。简单性不是单一维度——它总是相对于个人和情境而言的。
为何有效: 让行为更容易是行为设计中最可靠的策略。与动机不同,能力可以被系统地设计。你删除的每个字段、消除的每个步骤、设置的每个默认值,都会在福格行为模型中将行为向右移动,即使在低动机时也能跨越行动线。能力链提供了一个诊断工具:找到最薄弱的环节并修复它。
关键见解:
产品应用:
| 场景 | 应用 | 示例 |
|---|---|---|
| 注册 | 最小化所有六个因素的能力链成本 | 一键 SSO 消除了时间、脑力和非惯例因素 |
| 核心操作 | 找到并修复链中最薄弱的环节 | 如果脑力是瓶颈,则添加智能默认值和模板 |
| 移动端体验 | 针对体力消耗和时间限制进行优化 | 预填表单、拇指友好的目标、最少输入 |
| 企业采用 | 解决社会偏差和非惯例因素 | “您的团队已经在使用这个”降低了社会风险 |
| 摩擦审计 | 系统地测试六个因素中的每一个 | 为每个关键行为遍历每个因素并评分 1-5 |
文案模式:
伦理边界: 减少摩擦应该让真正有价值的行为更容易。切勿减少有害行为的摩擦(例如,让过度消费、过度分享或未经确认删除重要数据变得太容易)。
详细了解六个因素、摩擦审计模板和简化策略,请参阅:references/ability-chain.md。
核心概念: 提示是行动号召——它说“现在就做”。没有提示,无论动机和能力如何,行为都不会发生。有三种类型:个人提示(内部提醒)、情境提示(环境线索)和行动提示(产品设计的触发器)。
为何有效: 提示是最容易被忽视的要素。许多产品团队认为,如果动机和能力都存在,行为就会发生。但不会——没有适时的提示就不会。关键见解:提示只有在行动线以上才有效。向缺乏能力或动机的人发送推送通知是垃圾信息,而不是参与。最好的提示出现在动机和能力已经足够的时候。
关键见解:
产品应用:
| 场景 | 应用 | 示例 |
|---|---|---|
| 通知 | 仅在用户高于行动线时提示 | 当用户有内容需要查看时发送摘要,而不是按计划发送 |
| 新用户引导 | 使用行动提示来引导首次行为 | 工具提示:在正确时刻“点击此处创建您的第一个项目” |
| 习惯养成 | 设计基于锚点的提示 | “在您早上的站会之后,查看您的仪表板” |
| 重新参与 | 与现实事件相关的情境提示 | “您的报告已准备就绪”(基于事件,而非基于时间) |
| 功能发现 | 在动机和能力一致时提示 | 当用户遇到功能能解决的问题时,显示功能导览 |
文案模式:
伦理边界: 尊重提示疲劳和用户偏好。每个提示都应通过测试:“我现在会乐意收到这个吗?”切勿使用提示通过焦虑或制造的紧迫感来操纵用户。
了解提示类型、时机策略、通知设计和锚点时刻,请参阅:references/prompt-design.md。
微习惯方法是 B=MAP 的实际应用:让行为变得如此微小,几乎不需要动机,将它们锚定到现有惯例中,并立即庆祝。
After I [ANCHOR MOMENT], I will [TINY BEHAVIOR], then I [CELEBRATION].
锚点时刻: 一个可靠发生的现有惯例(打开应用、结束会议、早晨喝咖啡)。
微小行为: 目标行为的最小版本——小到几乎不可能跳过。不是“写一份报告”,而是“打开报告模板”。不是“查看分析”,而是“瞥一眼一个指标”。
庆祝: 行为后立即产生的积极情绪。庆祝能巩固习惯。没有它,仅靠重复是不够的——你需要成功的感觉。
每个目标行为都有一个启动步骤——最小但有意义的版本:
| 目标行为 | 启动步骤 | 为何有效 |
|---|---|---|
| 完成新用户引导 | 填写一个字段 | 完成带来的动力 |
| 每日使用分析 | 打开仪表板 | 看到数据会产生好奇心 |
| 与团队协作 | 发送一条评论 | 社会互惠开始起作用 |
| 编写文档 | 打开文档模板 | 消除了空白页阻力 |
| 查看每周指标 | 标记一个指标 | 创造个人投入感 |
一旦微小行为被巩固,它自然会增长:
关键:切勿强迫扩展。让动机和动力驱动扩展。微小版本不是失败——它是基础。
完整的微习惯配方、庆祝技巧和扩展模式,请参阅:references/tiny-habits.md。
福格用于创造持久行为改变的系统化流程:
用户想要什么结果?不是产品的目标——是用户的愿望。
列出所有可能实现愿望的行为。要详尽——先不要承诺一个行为。
对于每个行为,评估:用户是否有足够的动机?它是否足够容易?使用焦点映射技术——将行为绘制在影响与可行性的 2×2 矩阵上。
选取匹配度最高的行为,并将其缩小到其启动步骤。设计提示。添加庆祝。
一旦微小行为被巩固,就扩展它。使用能力链修复瓶颈。根据时机数据优化提示。
行动线是福格行为模型中的视觉阈值。在线以上,行为在提示时发生。在线以下,则不会。
两种可靠策略:
1. 提高能力(向右移动)
2. 寻找更好的提示(在正确时刻提示)
不可靠的策略:提高动机(向上移动)
将 B=MAP 映射到产品指标:
| 指标 | B=MAP 诊断 | 行动 |
|---|---|---|
| 低激活率 | 首次行动低于行动线 | 将新用户引导缩小到启动步骤;修复最薄弱的能力链环节 |
| 第 1 天流失 | 提示失败或时机不当 | 重新设计首日提示;锚定到现有用户惯例 |
| 第 7 天流失 | 动机浪潮退去,行为太难 | 降低核心操作难度;不要依赖初始兴奋感 |
| 第 30 天流失 | 习惯未形成,无内部提示 | 创建微习惯配方;添加庆祝/反馈循环 |
| 低功能采用率 | 功能对大多数用户来说低于行动线 | 对功能进行摩擦审计;仅在动机存在时提示 |
| 通知疲劳 | 在行动线以下发送提示 | 减少提示量;仅在用户有动机+能力时发送 |
| 错误 | 为何失败 | 修复方法 |
|---|---|---|
| 依赖动机驱动留存 | 动机是波浪——它总会退去。需要高动机的产品在低谷时会失败 | 为低动机时刻设计;让行为微小到足以在动机下降时存活 |
| 忽视能力链瓶颈 | 你优化了时间,但真正的障碍是脑力或社会偏差 | 审计所有六个因素;修复最稀缺的资源,而不是最明显的那个 |
| 在行动线以下发送提示 | 向缺乏能力和动机的用户发送推送通知是垃圾信息,不是参与 | 仅在动机+能力足够时提示;使用基于事件的触发器 |
| 在新用户引导中跳过庆祝 | 没有积极情绪,仅靠重复无法巩固习惯 | 在关键操作后添加即时反馈、成功状态和微庆祝 |
| 让首次行动过于宏大 | “完善您的个人资料”是一个项目,不是一个行为。用户在开始前就放弃了 | 缩小到启动步骤:“添加您的姓名”或“上传照片”——一个字段,一个操作 |
| 复制成功产品而不诊断 B=MAP | 如果能力或提示不同,对高动机受众有效的方法对你的受众可能失败 | 始终首先诊断你特定用户的动机、能力和提示情境 |
| 问题 | 如果答案为“否” | 行动 |
|---|---|---|
| 新用户能否在 60 秒内完成核心操作? | 能力太低 | 使用能力链进行摩擦审计;缩小到启动步骤 |
| 产品在用户动机低时是否有效? | 设计依赖动机峰值 | 重新设计核心行为,使其需要最小动机 |
| 提示是否与现实事件或锚点时刻相关? | 提示感觉像垃圾信息 | 从基于时间切换到基于事件或基于锚点的提示 |
| 关键操作后是否有即时反馈? | 没有庆祝 = 没有习惯巩固 | 添加成功状态、进度指示器或社会反馈 |
| 你是否识别了能力链中最薄弱的环节? | 你在优化错误的东西 | 为你的核心行为对六个因素中的每一个评分 1-5 |
| 用户是否自然地从微小行为扩展到全面参与? | 过早强迫复杂行为 | 实施启动步骤,让行为有机增长 |
此技能基于 BJ Fogg 开发的行为设计研究。关于完整的方法论、研究和案例研究:
BJ Fogg 博士 是斯坦福大学行为设计实验室的创始人,自 1998 年以来一直指导行为改变研究。他创建了福格行为模型(B=MAT/B=MAP),该模型已成为全球产品设计师、健康研究人员和行为改变专业人士使用的基础框架。福格创造了“行为设计”一词,并培训了数千名创新者使用他的方法,其中包括 Instagram 的创始人(Mike Krieger 曾是他的学生)。他对“说服性技术”(计算机作为说服性技术)的研究开创了一个全新的学术领域。《微习惯》将二十年的研究提炼成一个实用的行为改变系统,证明持久的改变并非来自动机或意志力,而是来自将行为设计得微小、有锚点并值得庆祝。福格的方法已被从硅谷初创公司到财富 500 强企业的产品团队采用,他的学术著作被引用超过 20,000 次。
每周安装次数
260
代码库
GitHub 星标数
260
首次出现
2026年2月19日
安全审计
安装于
codex246
opencode245
gemini-cli242
kimi-cli242
github-copilot242
amp242
Framework for designing products that reliably change behavior. Based on a fundamental truth: behavior is not about willpower or motivation—it is a design problem with a predictable equation.
The Fogg Behavior Model = B=MAP. Behavior happens when Motivation, Ability, and a Prompt converge at the same moment.
HIGH ┃
┃ ★ Behavior happens
┃ (above the Action Line)
┃
Motivation ┃━━━━━━━━━━━━━━━━━━━━━━━ ← Action Line
┃
┃ ✗ Behavior fails
┃ (below the Action Line)
LOW ┃
┗━━━━━━━━━━━━━━━━━━━━━━━━━
HARD EASY
Ability
The Action Line: When motivation and ability are sufficient, a prompt causes the behavior. Below the line, no prompt works. The line curves: high motivation compensates for low ability, and high ability compensates for low motivation. The reliable strategy is to make behaviors easier (move right), not to pump up motivation (move up).
Goal: 10/10. When reviewing or creating product behavior design, rate them 0-10 based on adherence to the principles below. A 10/10 means full alignment with all guidelines; lower scores indicate gaps to address. Always provide the current score and specific improvements needed to reach 10/10.
Core concept: Motivation is the energy for action. It has three core motivators, each with two sides: Sensation (pleasure/pain), Anticipation (hope/fear), and Belonging (acceptance/rejection). Motivation is powerful but unreliable — it fluctuates like waves.
Why it works: Motivation explains why people want to act, but it is the least reliable element in behavior design. It spikes (New Year's resolutions, product launches) and crashes (day 3, week 2). Products that depend on high motivation fail when the wave recedes. The best designs work even when motivation is at a trough.
Key insights:
Product applications:
| Context | Application | Example |
|---|---|---|
| Onboarding | Don't rely on new-user motivation spike lasting | Design first actions to work even when excitement fades |
| Re-engagement | Assume returning users have low motivation | Show immediate value before asking for effort |
| Feature adoption | Match motivation to difficulty | Simple features need low motivation; complex ones need motivation boosters |
| Messaging | Tap into the right motivator | Social fitness → belonging; financial tool → anticipation (hope) |
| Churn prevention | Diagnose if motivation dropped or was never high | Survey churned users for motivational misalignment |
Copy patterns:
Ethical boundary: Never manufacture false hope or exploit fear to inflate motivation. Motivation tactics should connect users to genuine outcomes, not create anxiety that drives compulsive usage.
See: references/motivation-waves.md for deep dive on the three motivators, motivation waves, and designing for troughs.
Core concept: Ability is the capacity to do the behavior. It is a function of the scarcest resource across six factors — the Ability Chain. If any single link is too weak (too expensive, too time-consuming, too confusing), the behavior breaks. Simplicity is not a single dimension — it is always relative to the person and context.
Why it works: Making behaviors easier is the most reliable strategy in behavior design. Unlike motivation, ability can be systematically engineered. Every field you remove, every step you eliminate, every default you set moves the behavior to the right on the Fogg Behavior Model, crossing the Action Line even at low motivation. The Ability Chain provides a diagnostic: find the weakest link and fix it.
Key insights:
Product applications:
| Context | Application | Example |
|---|---|---|
| Signup | Minimize Ability Chain cost across all six factors | One-click SSO eliminates time, mental effort, and non-routine |
| Core action | Find and fix the weakest link in the chain | If mental effort is the bottleneck, add smart defaults and templates |
| Mobile experience | Optimize for physical effort and time constraints | Pre-filled forms, thumb-friendly targets, minimal typing |
| Enterprise adoption | Address social deviance and non-routine factors | "Your team already uses this" reduces social risk |
| Friction audit | Systematically test each of the six factors | Walk through each factor for every key behavior and rate 1-5 |
Copy patterns:
Ethical boundary: Reducing friction should make genuinely valuable behaviors easier. Never reduce friction on harmful actions (e.g., making it too easy to overspend, over-share, or delete important data without confirmation).
See: references/ability-chain.md for the six factors in detail, friction audit templates, and simplification strategies.
Core concept: The prompt is the call to action — the thing that says "do it now." Without a prompt, behavior doesn't happen regardless of motivation and ability. Three types: Person Prompts (internal reminders), Context Prompts (environmental cues), and Action Prompts (designed triggers from the product).
Why it works: Prompts are the most overlooked element. Many product teams assume that if motivation and ability are present, behavior will happen. It won't — not without a well-timed prompt. The key insight: prompts only work above the Action Line. Sending a push notification to someone who lacks ability or motivation is spam. The best prompts arrive when motivation and ability are already sufficient.
Key insights:
Product applications:
| Context | Application | Example |
|---|---|---|
| Notifications | Only prompt when user is above the Action Line | Send digest when user has content to review, not on a schedule |
| Onboarding | Use action prompts to guide first behaviors | Tooltip: "Click here to create your first project" at the right moment |
| Habit formation | Design anchor-based prompts | "After your morning standup, review your dashboard" |
| Re-engagement | Context prompts tied to real events | "Your report is ready" (event-based, not time-based) |
| Feature discovery | Prompt when motivation and ability align | Show feature tour when user encounters the problem it solves |
Copy patterns:
Ethical boundary: Respect prompt fatigue and user preferences. Every prompt should pass the test: "Would I appreciate receiving this right now?" Never use prompts to manipulate through anxiety or manufactured urgency.
See: references/prompt-design.md for prompt types, timing strategies, notification design, and anchor moments.
The Tiny Habits method is the practical application of B=MAP: make behaviors so small they require almost no motivation, anchor them to existing routines, and celebrate immediately.
After I [ANCHOR MOMENT], I will [TINY BEHAVIOR], then I [CELEBRATION].
Anchor Moment: An existing routine that reliably happens (opening an app, finishing a meeting, morning coffee).
Tiny Behavior: The smallest version of the target behavior — so small it's almost impossible to skip. Not "write a report" but "open the report template." Not "review analytics" but "glance at one metric."
Celebration: An immediate positive emotion after doing the behavior. Celebration wires the habit. Without it, repetition alone isn't enough — you need the feeling of success.
Every target behavior has a Starter Step — the tiniest meaningful version:
| Target Behavior | Starter Step | Why It Works |
|---|---|---|
| Complete onboarding | Fill in one field | Momentum from completion |
| Use analytics daily | Open the dashboard | Seeing data creates curiosity |
| Collaborate with team | Send one comment | Social reciprocity kicks in |
| Write documentation | Open the doc template | Blank page resistance removed |
| Review weekly metrics | Star one metric | Creates personal investment |
Once the tiny behavior is wired, it naturally grows:
The key: never force scaling. Let motivation and momentum drive expansion. The tiny version is not a failure — it is the foundation.
See: references/tiny-habits.md for the full Tiny Habits recipe, celebration techniques, and scaling patterns.
Fogg's systematic process for creating lasting behavior change:
What outcome does the user want? Not the product's goal — the user's aspiration.
List all possible behaviors that could achieve the aspiration. Be exhaustive — don't commit to one behavior yet.
For each behavior, assess: Does the user have enough motivation? Is it easy enough? Use the Focus Mapping technique — plot behaviors on a 2×2 of impact vs. feasibility.
Take the best-matched behavior and shrink it to its Starter Step. Design the prompt. Add celebration.
Once the tiny behavior is wired, expand it. Fix bottlenecks using the Ability Chain. Refine prompts based on timing data.
The Action Line is the visual threshold in the Fogg Behavior Model. Above it, behaviors happen when prompted. Below it, they don't.
Two reliable strategies:
1. Increase Ability (move right)
2. Find better Prompts (prompt at the right moment)
Unreliable strategy: Increase Motivation (move up)
Map B=MAP to product metrics:
| Metric | B=MAP Diagnosis | Action |
|---|---|---|
| Low activation | First action is below the Action Line | Shrink onboarding to Starter Step; fix weakest Ability Chain link |
| Day-1 drop-off | Prompt failed or mistimed | Redesign first-day prompts; anchor to existing user routine |
| Day-7 drop-off | Motivation wave receded, behavior too hard | Reduce core action difficulty; don't depend on initial excitement |
| Day-30 drop-off | Habit didn't form, no internal prompt | Create tiny habit recipe; add celebration/feedback loops |
| Low feature adoption | Feature is below Action Line for most users | Friction audit the feature; prompt only when motivation is present |
| Notification fatigue | Prompts sent below the Action Line | Reduce prompt volume; send only when user has motivation + ability |
| Mistake | Why It Fails | Fix |
|---|---|---|
| Relying on motivation to drive retention | Motivation is a wave — it always recedes. Products that need high motivation fail at the trough | Design for low-motivation moments; make behaviors tiny enough to survive motivation dips |
| Ignoring the Ability Chain bottleneck | You optimized time but the real barrier is mental effort or social deviance | Audit all six factors; fix the scarcest resource, not the most obvious one |
| Sending prompts below the Action Line | Push notifications to unmotivated users who lack ability is spam, not engagement | Only prompt when motivation + ability are sufficient; use event-based triggers |
| Skipping celebration in onboarding | Without positive emotion, repetition alone doesn't wire habits | Add immediate feedback, success states, and micro-celebrations after key actions |
| Making the first action too ambitious | "Complete your profile" is a project, not a behavior. Users abandon before starting | Shrink to Starter Step: "Add your name" or "Upload a photo" — one field, one action |
| Copying successful products without diagnosing B=MAP |
| Question | If No | Action |
|---|---|---|
| Can a new user complete the core action in under 60 seconds? | Ability is too low | Friction audit using the Ability Chain; shrink to Starter Step |
| Does the product work when user motivation is low? | Design depends on motivation spikes | Redesign core behaviors to require minimal motivation |
| Are prompts tied to real events or anchor moments? | Prompts feel like spam | Switch from time-based to event-based or anchor-based prompts |
| Is there immediate feedback after key actions? | No celebration = no habit wiring | Add success states, progress indicators, or social feedback |
| Have you identified the weakest link in the Ability Chain? | You're optimizing the wrong thing | Rate each of the six factors 1-5 for your core behavior |
| Do users naturally scale from tiny behaviors to full engagement? | Forcing complex behaviors too early | Implement Starter Steps and let behaviors grow organically |
This skill is based on the behavior design research developed by BJ Fogg. For the complete methodology, research, and case studies:
BJ Fogg, PhD is the founder of the Behavior Design Lab at Stanford University, where he has directed research on behavior change since 1998. He created the Fogg Behavior Model (B=MAT/B=MAP), which has become the foundational framework used by product designers, health researchers, and behavior change professionals worldwide. Fogg coined the term "behavior design" and has trained thousands of innovators in his methods, including the founders of Instagram (Mike Krieger was a student). His research on "captology" (computers as persuasive technology) created an entirely new academic discipline. Tiny Habits distills two decades of research into a practical system for behavior change, demonstrating that lasting change comes not from motivation or willpower but from designing behaviors to be tiny, anchored, and celebrated. Fogg's methods have been adopted by product teams at companies from Silicon Valley startups to Fortune 500 enterprises, and his academic work has been cited over 20,000 times.
Weekly Installs
260
Repository
GitHub Stars
260
First Seen
Feb 19, 2026
Security Audits
Gen Agent Trust HubPassSocketPassSnykPass
Installed on
codex246
opencode245
gemini-cli242
kimi-cli242
github-copilot242
amp242
注册流程转化率优化指南:减少摩擦、提高完成率的专家技巧
25,700 周安装
| What works for a high-motivation audience fails for yours if ability or prompts differ |
| Always diagnose your specific users' motivation, ability, and prompt context first |