重要前提
安装AI Skills的关键前提是:必须科学上网,且开启TUN模式,这一点至关重要,直接决定安装能否顺利完成,在此郑重提醒三遍:科学上网,科学上网,科学上网。查看完整安装教程 →
cognitive-biases by guia-matthieu/clawfu-skills
npx skills add https://github.com/guia-matthieu/clawfu-skills --skill cognitive-biases将卡尼曼和泰勒的行为经济学原理应用于营销——理解客户如何实际做出决策,并设计与人类心理学协同而非对抗的体验。
当您需要时使用此技能:
此技能对以下人员尤其有价值:
来源: 丹尼尔·卡尼曼(《思考,快与慢》)& 理查德·泰勒(《助推》、《错误的行为》)
核心原则: 人类不像理性的经济主体那样做决策。我们使用导致可预测偏差的心理捷径(启发法)。理解这些偏差可以让您设计与人类心理学协同而非对抗的营销。
"如果你想鼓励某人做某事,就让它变得容易。" — 理查德·泰勒
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
| Claude 做什么 | 您决定什么 |
|---|---|
| 构建内容框架 | 最终信息传递 |
| 建议说服技巧 | 品牌声音 |
| 创建草稿变体 | 版本选择 |
| 识别优化机会 | 发布时间 |
| 分析竞争对手方法 | 战略方向 |
当调用时,我将指导您将认知偏差应用于营销:
提供关于您营销挑战的信息:
示例提示:
有帮助的信息:
所有人类决策都通过两个心理系统进行:
| 系统 1(快) | 系统 2(慢) |
|---|---|
| 自动的 | 深思熟虑的 |
| 无意识的 | 有意识的 |
| 不费力的 | 费力的 |
| 情感的 | 理性的 |
| 直觉的 | 分析的 |
| 始终在线 | 容量有限 |
关键洞察: 大多数购买决策是系统 1 的。您的营销必须诉诸直觉和情感,而不仅仅是逻辑。
应用:
为系统 1 设计,提供系统 2 的理由。
1. 损失厌恶 损失的感觉比同等收益带来的愉悦感强烈约 2 倍。
| 与其... | 尝试... |
|---|---|
| "节省 50 美元" | "不要损失 50 美元" |
| "获得这些功能" | "不要错过这些功能" |
| "注册获取更新" | "不要错过重要更新" |
应用:
2. 锚定效应 看到的第一个数字会影响所有后续判断。
应用:
示例:
"原价 299 美元" → "149 美元"(锚定在 299 美元)
对比
"149 美元"(无锚定,感觉昂贵)
3. 诱饵效应 第三个较差的选项使目标选项看起来更好。
经典示例:
添加诱饵:
定价页面应用: 设计三个层级,其中中间层级是目标。"企业版" 层级使 "专业版" 看起来合理。
4. 框架效应 相同的信息,不同的呈现方式 = 不同的决策。
| 负面框架 | 正面框架 |
|---|---|
| 10% 脂肪 | 90% 无脂肪 |
| 5% 失败率 | 95% 成功率 |
| 每天花费 10 美元 | 每天节省价值 50 美元的时间 |
应用: 将利益框架化为收益,将竞争框架化为您所避免的。
5. 稀缺性偏差 有限的可用性增加感知价值。
稀缺性类型:
| 类型 | 示例 |
|---|---|
| 数量 | "仅剩 3 件库存" |
| 时间 | "优惠 24 小时后结束" |
| 访问权限 | "仅限会员价格" |
| 信息 | "独家内部报告" |
警告: 虚假的稀缺性会破坏信任。使用真实的限制条件。
6. 当下偏见 即时回报胜过未来回报。
应用:
7. 社会认同 人们跟随他人的行为,尤其是在不确定的情况下。
社会认同类型:
| 类型 | 示例 | 最适合 |
|---|---|---|
| 数字 | "50,000+ 客户" | 规模、信任 |
| 推荐语 | "改变了我的业务" | 情感连接 |
| 徽标 | "受 Nike、Google 信赖" | B2B 可信度 |
| 评论 | "来自 2,000 条评论的 4.8/5 分" | 消费品 |
| 活动 | "来自纽约市的 John 刚刚购买" | 错失恐惧症、活动 |
| 专家 | "医生推荐" | 权威性 |
最佳实践: 使社会认同与客户的关注点相匹配。
8. 从众/羊群行为 在不确定的情况下,人们跟随大众。
应用:
9. 认知超载 选项太多 = 决策瘫痪 = 无行动。
果酱研究: 24 种果酱口味 → 3% 购买。6 种口味 → 30% 购买。
应用:
10. 现状偏见 人们倾向于不改变,即使改变是有益的。
应用:
示例: 退出式器官捐赠与加入式相比,参与率有显著差异。
11. 目标梯度效应 人们在接近完成时加速努力。
应用:
示例: 预先盖好 10 个章中 2 个的忠诚卡,优于空白的 8 个章的卡。
12. 禀赋效应 人们高估他们已经拥有的东西。
应用:
13. 承诺与一致性 人们的行为与先前的承诺保持一致。
应用:
14. 乐观偏见 人们高估自己的积极结果。
应用:
15. 可得性偏差 生动、近期的例子感觉更可能发生。
应用:
16. 信息厌恶(鸵鸟效应) 人们回避他们担心会是负面的信息。
应用:
使用英国行为洞察团队的框架来构建干预措施:
| 原则 | 应用 |
|---|---|
| 简单 | 减少摩擦、简化、使用默认选项 |
| 有吸引力 | 使其视觉上吸引人、个性化、使用激励 |
| 社交性 | 展示他人做什么、使用社交承诺 |
| 及时性 | 在决策点进行干预、使用提示 |
道德使用原则:
红线:
当前问题: 访客查看定价页面但不转化。大多数人选择免费套餐。
行为诊断:
重新设计的定价页面:
热门 ↓
┌─────────────┬─────────────┬─────────────┐
│ 基础版 │ 专业版 │ 企业版 │
│ $29/月 │ $79/月 │ $199/月 │
├─────────────┼─────────────┼─────────────┤
│ 5 个项目 │ 25 个项目 │ 无限制 │
│ 1 个用户 │ 5 个用户 │ 无限制 │
│ 电子邮件 │ 优先支持 │ 专属 │
│ │ 集成 │ 自定义 API │
│ │ 分析 │ SLA │
└─────────────┴─────────────┴─────────────┘
"受 5,000+ 团队信赖"
[开始免费试用 →]
"30 天免费试用,无需信用卡"
应用的偏差:
附加元素:
当前问题: 70% 的购物车放弃率
行为诊断:
干预措施:
1. 解决损失厌恶:
2. 减少认知负荷:
3. 添加社会认同:
4. 创造紧迫性:
5. 承诺机制:
放弃购物车邮件序列:
邮件 1(1 小时后):"您遗漏了某些东西" + 损失厌恶("不要错过")
邮件 2(24 小时后):社会认同("顾客喜欢这个")+ 稀缺性("热卖中")
邮件 3(72 小时后):激励 + 紧迫性("10% 折扣明天到期")
对于任何营销资产,检查您利用了哪些偏差:
定价与优惠:
文案与信息传递:
用户体验:
页面/流程:_______________
目标:_______________
当前状态:
- 转化率:____%
- 主要流失点:_______________
- 用户反馈:_______________
行为障碍:
1. [ ] 认知超载(选项/步骤太多)
2. [ ] 损失厌恶(意外成本、不确定性)
3. [ ] 现状偏见(没有令人信服的理由行动)
4. [ ] 缺少社会认同(没有验证)
5. [ ] 无紧迫性(可以稍后决定)
6. [ ] 其他:_______________
待测试的干预措施:
障碍:_______________
利用的偏差:_______________
具体更改:_______________
假设:_______________
障碍:_______________
利用的偏差:_______________
具体更改:_______________
假设:_______________
测量:
- 主要指标:_______________
- 次要指标:_______________
- 测试时长:_______________
| 偏差 | 公式 | 示例 |
|---|---|---|
| 稀缺性 | "仅剩 X:[利益]" | "仅剩 3 个名额:加入大师班" |
| 损失厌恶 | "不要错过:[机会]" | "不要错过:您的折扣今晚到期" |
| 社会认同 | "[数字] 人 [行动]" | "5,000 名营销人员阅读了此通讯" |
| 好奇心缺口 | "[引人入胜的陈述] 原因如下" | "我们差点没发布这个。原因如下。" |
| 个性化 | "[姓名],您的 [物品] 在等待" | "Sarah,您的购物车在等待" |
| 紧迫性 | "[时间]:[需要采取的行动]" | "24 小时:完成您的申请" |
主要来源:
额外资源:
每周安装数
51
仓库
GitHub 星标数
50
首次出现
2026年2月13日
安全审计
安装于
opencode50
gemini-cli49
codex49
cursor48
github-copilot48
amp46
Apply Kahneman and Thaler's behavioral economics principles to marketing—understand how customers actually make decisions and design experiences that work with human psychology, not against it.
Use this skill when you need to:
This skill is particularly valuable for:
Source: Daniel Kahneman (Thinking, Fast and Slow) & Richard Thaler (Nudge , Misbehaving)
Core Principle: Humans don't make decisions like rational economic agents. We use mental shortcuts (heuristics) that lead to predictable biases. Understanding these biases allows you to design marketing that works with human psychology rather than against it.
"If you want to encourage someone to do something, make it easy." — Richard Thaler
| Claude Does | You Decide |
|---|---|
| Structures content frameworks | Final messaging |
| Suggests persuasion techniques | Brand voice |
| Creates draft variations | Version selection |
| Identifies optimization opportunities | Publication timing |
| Analyzes competitor approaches | Strategic direction |
When invoked, I will guide you through applying cognitive biases to marketing:
Provide information about your marketing challenge:
Example prompts:
Information that helps:
All human decisions flow through two mental systems:
| System 1 (Fast) | System 2 (Slow) |
|---|---|
| Automatic | Deliberate |
| Unconscious | Conscious |
| Effortless | Effortful |
| Emotional | Rational |
| Intuitive | Analytical |
| Always on | Limited capacity |
Key Insight: Most purchase decisions are System 1. Your marketing must appeal to intuition and emotion, not just logic.
Application:
Design for System 1, provide System 2 justification.
1. Loss Aversion Losses feel ~2x more painful than equivalent gains feel good.
| Instead of... | Try... |
|---|---|
| "Save $50" | "Don't lose $50" |
| "Get these features" | "Don't miss these features" |
| "Sign up for updates" | "Don't miss important updates" |
Applications:
2. Anchoring Effect First number seen influences all subsequent judgments.
Applications:
Example:
"Was $299" → "$149" (anchored to $299)
vs.
"$149" (no anchor, feels expensive)
3. Decoy Effect A third inferior option makes target option look better.
Classic Example:
Add decoy:
Pricing Page Application: Design three tiers where middle tier is the target. The "Enterprise" tier makes "Pro" look reasonable.
4. Framing Effect Same information, different presentation = different decisions.
| Negative Frame | Positive Frame |
|---|---|
| 10% fat | 90% fat-free |
| 5% failure rate | 95% success rate |
| Costs $10/day | Saves you time worth $50/day |
Application: Frame benefits as gains, competition as what you avoid.
5. Scarcity Bias Limited availability increases perceived value.
Types of Scarcity:
| Type | Example |
|---|---|
| Quantity | "Only 3 left in stock" |
| Time | "Offer ends in 24 hours" |
| Access | "Members-only pricing" |
| Information | "Exclusive insider report" |
Warning: Fake scarcity destroys trust. Use real constraints.
6. Present Bias Immediate rewards trump future rewards.
Applications:
7. Social Proof People follow what others do, especially in uncertainty.
Types of Social Proof:
| Type | Example | Best For |
|---|---|---|
| Numbers | "50,000+ customers" | Scale, trust |
| Testimonials | "Changed my business" | Emotional connection |
| Logos | "Trusted by Nike, Google" | B2B credibility |
| Reviews | "4.8/5 from 2,000 reviews" | Consumer products |
| Activity | "John from NYC just purchased" | FOMO, activity |
| Expert | "Recommended by doctors" | Authority |
Best Practice: Match social proof to your customer's concerns.
8. Conformity/Herd Behavior In uncertainty, people follow the crowd.
Applications:
9. Cognitive Overload Too many options = decision paralysis = no action.
The Jam Study: 24 jam flavors → 3% bought. 6 flavors → 30% bought.
Applications:
10. Status Quo Bias People prefer not to change, even when change is beneficial.
Applications:
Example: Opt-out organ donation vs. opt-in shows dramatic difference in participation.
11. Goal Gradient Effect People accelerate effort as they near completion.
Applications:
Example: Loyalty card with 2 of 10 stamps pre-filled outperforms blank 8-stamp card.
12. Endowment Effect People overvalue what they already possess.
Applications:
13. Commitment & Consistency People act consistently with prior commitments.
Applications:
14. Optimism Bias People overestimate positive outcomes for themselves.
Applications:
15. Availability Bias Vivid, recent examples feel more likely.
Applications:
16. Information Aversion (Ostrich Effect) People avoid information they fear will be negative.
Applications:
Use the UK Behavioural Insights Team's framework to structure interventions:
| Principle | Application |
|---|---|
| Easy | Reduce friction, simplify, use defaults |
| Attractive | Make it visually appealing, personalize, use incentives |
| Social | Show what others do, use social commitments |
| Timely | Intervene at decision points, use prompts |
Principles for Ethical Use:
Red Lines:
Current Problem: Visitors view pricing page but don't convert. Most choose Free tier.
Behavioral Diagnosis:
Redesigned Pricing Page:
POPULAR ↓
┌─────────────┬─────────────┬─────────────┐
│ BASIC │ PRO │ ENTERPRISE │
│ $29/mo │ $79/mo │ $199/mo │
├─────────────┼─────────────┼─────────────┤
│ 5 projects │ 25 projects │ Unlimited │
│ 1 user │ 5 users │ Unlimited │
│ Email │ Priority │ Dedicated │
│ │ Integrations│ Custom API │
│ │ Analytics │ SLA │
└─────────────┴─────────────┴─────────────┘
"Trusted by 5,000+ teams"
[Start Free Trial →]
"30-day free trial, no credit card"
Biases Applied:
Additional Elements:
Current Problem: 70% cart abandonment rate
Behavioral Diagnosis:
Interventions:
1. Address Loss Aversion:
2. Reduce Cognitive Load:
3. Add Social Proof:
4. Create Urgency:
5. Commitment Devices:
Abandoned Cart Email Sequence:
Email 1 (1 hour): "You left something behind" + loss aversion ("Don't miss out") Email 2 (24 hours): Social proof ("Customers love this") + scarcity ("Selling fast") Email 3 (72 hours): Incentive + urgency ("10% off expires tomorrow")
For any marketing asset, check which biases you're leveraging:
Pricing & Offers:
Copy & Messaging:
User Experience:
PAGE/FLOW: _______________
GOAL: _______________
CURRENT STATE:
- Conversion rate: ____%
- Main drop-off point: _______________
- User feedback: _______________
BEHAVIORAL BARRIERS:
1. [ ] Cognitive overload (too many options/steps)
2. [ ] Loss aversion (surprise costs, uncertainty)
3. [ ] Status quo bias (no compelling reason to act)
4. [ ] Missing social proof (no validation)
5. [ ] No urgency (can decide later)
6. [ ] Other: _______________
INTERVENTIONS TO TEST:
Barrier: _______________
Bias to leverage: _______________
Specific change: _______________
Hypothesis: _______________
Barrier: _______________
Bias to leverage: _______________
Specific change: _______________
Hypothesis: _______________
MEASUREMENT:
- Primary metric: _______________
- Secondary metrics: _______________
- Test duration: _______________
| Bias | Formula | Example |
|---|---|---|
| Scarcity | "Only X left: [benefit]" | "Only 3 spots left: Join the masterclass" |
| Loss aversion | "Don't miss: [opportunity]" | "Don't miss: Your discount expires tonight" |
| Social proof | "[Number] people [action]" | "5,000 marketers read this newsletter" |
| Curiosity gap | "[Intriguing statement] Here's why" | "We almost didn't launch this. Here's why." |
| Personalization | "[Name], your [item] is waiting" | "Sarah, your cart is waiting" |
| Urgency | "[Time]: [action required]" | "24 hours: Complete your application" |
Primary Sources:
Additional Resources:
Weekly Installs
51
Repository
GitHub Stars
50
First Seen
Feb 13, 2026
Security Audits
Gen Agent Trust HubPassSocketPassSnykPass
Installed on
opencode50
gemini-cli49
codex49
cursor48
github-copilot48
amp46
normalize设计标准化工具:前端UI组件与设计系统一致性优化
54,300 周安装