influence-psychology
정보
이 스킬은 제품 디자인과 마케팅 문구에 시알디니의 윤리적 설득 7원칙을 적용하기 위한 프레임워크를 제공합니다. 사용자가 전환율 개선, 신뢰 신호 강화 또는 체험담 구성이나 긴박성 전략과 같은 설득적 메시지 작성이 필요할 때 작동합니다. 이 가이드는 각 원칙의 적용 시기와 윤리적 한계를 다룹니다.
빠른 설치
Claude Code
추천npx skills add wondelai/skills -a claude-code/plugin add https://github.com/wondelai/skillsgit clone https://github.com/wondelai/skills.git ~/.claude/skills/influence-psychologyClaude Code에서 이 명령을 복사하여 붙여넣어 스킬을 설치하세요
문서
Influence Psychology Framework
Apply six decades of persuasion science — Cialdini's research into why people say "yes" — to product, copy, and sales, ethically.
Core Principle
People don't make decisions rationally — they use mental shortcuts (heuristics) that can be triggered to influence behavior. These shortcuts evolved because they're usually reliable, but they can also be exploited. Understanding them lets you design products, messaging, and experiences that align with how people actually decide.
Scoring
Goal: 10/10. When reviewing or creating persuasive elements (features, copy, flows, campaigns), run the Quick Diagnostic, then score against the bands below and apply the ethics gate. Always report the current score and the specific change needed to reach 10/10.
- 9-10 — Multiple principles deliberately layered; every claim truthful; users can reverse the decision; passes the transparency test (still works if the user knows the strategy); safe for vulnerable users.
- 7-8 — Principles deliberately layered and honest, but one gap (e.g. weak reversibility, or a single principle where layering was possible).
- 5-6 — One principle present but generic, or leverage left on the table.
- <=3 — No principle deliberately designed (relying on luck), OR any tactic is deceptive/coercive. Any fabricated proof, fake scarcity, or hidden-cost dark pattern caps the score at 3 regardless of other strengths.
The Seven Principles of Influence
1. Reciprocity
Core concept: People feel obligated to give back to those who have given to them first.
Why it works: Humans are wired to avoid being indebted — the obligation to repay can overpower personal preference, and the return favor often exceeds the original gift.
Key insights:
- The gift must come first (before the request)
- Unexpected, personalized gifts beat expected, generic ones
- Even small gifts create obligation
Product applications:
| Context | Reciprocity Trigger | Example |
|---|---|---|
| Free trials | Full access first, then ask to pay | Spotify Premium trial → subscription |
| Content marketing | Value upfront (guides, tools) | HubSpot free CRM → paid tools |
| Referral programs | Reward both referrer and referee | Dropbox: both get extra storage |
Copy patterns:
- "Here's a gift for you..." (before asking)
- "As a thank you for signing up..."
- "We noticed you needed help with X, so we..."
See references/reciprocity.md when building free trials, lead magnets, or referral rewards — gift tiers, email templates, day-by-day reciprocity-stacking, and A/B variables.
2. Commitment & Consistency
Core concept: People want to be consistent with their past statements, beliefs, and actions.
Why it works: Inconsistency is psychologically uncomfortable; once we take a stand, personal and interpersonal pressure pushes us to behave consistently with it.
Key insights:
- Small initial commitments lead to larger ones (foot-in-the-door)
- Public > private; written > verbal; active (user-generated) > passive
- Self-perception: we infer our attitudes from our behavior
Product applications:
| Context | Commitment Trigger | Example |
|---|---|---|
| Onboarding | Easy yes, then larger asks | Duolingo: "Can you commit to 5 min/day?" |
| Goal setting | User publicly states a goal | Strava: "I want to run 50km this month" |
| Habit formation | Track streaks publicly | Snapchat streaks, GitHub contributions |
Copy patterns:
- "What's your biggest challenge with X?" (commitment to a problem)
- "How much would you like to save per month?" (numerical commitment)
- "You said you wanted to achieve X. Let's start with..."
Onboarding sequence: micro-commitment ("What brings you here?") → small action (click, choice) → public/written commitment (goal) → reinforce ("Based on what you told us...").
Ethical boundary: Make every commitment freely chosen and easily reversible — no trick opt-ins or locked-in defaults the user can't undo.
See references/commitment-consistency.md when designing onboarding or goal-setting flows — foot-in-the-door sequences and public-commitment tactics.
3. Social Proof
Core concept: People determine what's correct by finding out what others think is correct.
Why it works: When uncertain, we use others' behavior as a guide — "if everyone's doing it, it must be right."
Key insights:
- Most powerful when observers are uncertain; similar others = stronger proof
- Negative social proof backfires ("9 out of 10 don't...")
- Specific numbers beat vague claims ("2,347 users" > "thousands")
Types of social proof:
| Type | Definition | Example |
|---|---|---|
| Wisdom of crowds | Many people use/buy | "Join 50,000+ marketers" |
| Wisdom of friends | People you know use it | "3 of your friends use Notion" |
| Expert | Authorities endorse | "Recommended by Y Combinator" |
| Celebrity | Famous people use it | "Used by Elon Musk" |
| Certification | Third-party validation | "SOC 2 compliant", "App of the Year" |
| User | Similar people succeeded | "Startups like yours grew 10x" |
Product applications:
| Context | Social Proof Implementation | Example |
|---|---|---|
| Landing pages | User count, reviews, logos | "Trusted by 10,000+ companies" |
| Signup flow | Live signups, popular plans | "23 people signed up in the last hour" |
| Feature adoption | Show usage by others | "85% of teams use this feature" |
Copy patterns:
- "[X number] of [similar people] are already..."
- "[Name/Company] increased [metric] by [%]"
- "Don't take our word for it. Here's what [users] say..."
Ethical boundary: Disclose when proof is curated or cherry-picked (e.g. "selected reviews") rather than presenting it as representative.
See references/social-proof.md when building testimonial sections or trust bars — proof types and implementation patterns.
4. Authority
Core concept: People follow the lead of credible, knowledgeable experts.
Why it works: Obedience to authority is deeply ingrained — following experts is an efficient shortcut when we lack expertise ourselves.
Key insights:
- Titles, credentials, even symbols (lab coats, official-looking design) trigger automatic compliance
- Admitting a weakness paradoxically increases authority (trustworthiness) — lead with it before strengths
- Expertise doesn't transfer across domains, but people assume it does
Sources of authority:
| Type | Signal | Example |
|---|---|---|
| Credentials | Degrees, certifications | "Built by Stanford PhDs" |
| Experience | Years in field, track record | "20 years in cybersecurity" |
| Association | Trusted partners, investors | "Backed by Y Combinator" |
| Content | Thought leadership, research | "Based on research with 10,000 users" |
| Transparency | Honest about limitations | "Works best for teams of 10-50" |
Product applications:
| Context | Authority Trigger | Example |
|---|---|---|
| About page | Founder and team expertise | "Built by ex-Google engineers" |
| Content | Original research, citations | "State of [Industry] 2026 Report" |
| Partnerships | Security certs, integration badges | "SOC 2 Type II", "GDPR compliant" |
Copy patterns:
- "Trusted by [authority figure/company]"
- "Research shows that [cite source]..."
- "Our team includes [credentials]"
See references/authority.md when writing About pages, bylines, or trust badges — credential framing and thought-leadership strategies.
5. Liking
Core concept: People prefer to say yes to those they like.
Why it works: Each liking factor (similarity, compliments, cooperation) is a separate lever that independently lowers a person's resistance to a request — they stack.
Factors that increase liking:
| Factor | Mechanism | Example |
|---|---|---|
| Attractiveness | Halo effect: attractive = good | Professional headshots, polished design |
| Similarity | We like people like us | "I'm a founder just like you" |
| Compliments | Flattery works (even when obvious) | "You have great taste in tools" |
| Cooperation | Working toward shared goals | "Let's build this together" |
| Familiarity | Repeated exposure increases liking | Consistent brand, retargeting |
| Association | Linked to positive things | Placement with aspirational lifestyles |
Product applications:
| Context | Liking Trigger | Example |
|---|---|---|
| Brand voice | Friendly, conversational, human | Mailchimp's playful copy |
| Team pages | Real people, personality | Personal bios, hobbies, photos |
| Support | Warm, empathetic responses | "I totally understand that frustration..." |
Copy patterns:
- "We're [similar trait] just like you"
- "We built this because we were frustrated with..."
- Casual, warm language ("Hey", "Awesome!", "We got you")
See references/liking.md when setting brand voice or writing support replies — liking factors and tone guidelines. For ready-to-adapt persuasive copy across all seven principles, see references/copywriting.md.
6. Scarcity
Core concept: People want more of what they can't have or what's running out.
Why it works: Loss aversion is stronger than gain seeking — FOMO triggers urgency, and psychological reactance makes us want what threatens to become unavailable.
Key insights:
- Scarcity of time > scarcity of quantity; newly scarce > always scarce (loss framing)
- Competition increases value — if others want it, I want it
- Exclusive access is more valuable than open access
Types of scarcity:
| Type | Mechanism | Example |
|---|---|---|
| Limited quantity | Finite supply | "Only 5 seats left" |
| Limited time | Deadline pressure | "Offer ends Friday" |
| Exclusive access | Not everyone can have it | "Invite-only beta" |
| Competition | Others competing for it | "12 people viewing this" |
Product applications:
| Context | Scarcity Trigger | Example |
|---|---|---|
| Pricing | Limited-time discount | "Early bird pricing ends in 3 days" |
| Features | Beta access, waitlist | "Join 5,000 on the waitlist" |
| Inventory | Stock levels | "2 left in stock" |
Copy patterns:
- "Limited to the first [X] customers"
- "Offer expires [specific date]"
- "[X] people are viewing this right now"
Ethical boundary: Ethical scarcity reflects real constraints (true inventory counts, genuine deadlines, legitimate capacity limits). Unethical: invented limits, countdown timers that reset, "Only 2 left!" shown daily, pressuring vulnerable users.
See references/scarcity.md when adding urgency or waitlists — five scarcity types, each with an "ethical when" line and the dark patterns to avoid.
7. Unity
Core concept: People say yes to those they consider part of "us" (shared identity).
Why it works: Tribal identity is fundamental — we make sacrifices for in-group members we wouldn't make for strangers.
Unity vs. Liking: Liking = "this person is like me" (similarity); Unity = "this person is me" (shared identity).
Sources of unity:
| Type | Mechanism | Example |
|---|---|---|
| Place | Hometown, region, nationality | "Built in San Francisco, for founders" |
| Experience | Shared hardship or triumph | "We've all struggled with bad CRMs" |
| Values | Deep beliefs, mission alignment | "For people who value privacy" |
| Tribe | Co-creation, movement | "Join the indie maker community" |
Product applications:
| Context | Unity Trigger | Example |
|---|---|---|
| Brand positioning | Define the tribe | "For remote-first teams" |
| Messaging | "We" language, shared struggle | "We believe work should be flexible" |
| Community | Facilitate co-creation | User-generated content, forums |
Copy patterns:
- "For [identity group]" ("For designers", "For bootstrappers")
- "Join [X] others who believe..."
- "We're building this together"
Ethical boundary: Build the in-group by what it stands for, not by vilifying an out-group — define "us" without manufacturing a "them" to resent.
See references/unity.md when defining a brand tribe or community — identity-marketing and co-creation strategies.
Combining Principles
The most powerful persuasion layers multiple principles:
SaaS landing page: authority ("Built by ex-Stripe engineers") + social proof ("Trusted by 5,000+ companies") + liking (warm copy) + scarcity ("Join the beta—limited spots") + reciprocity ("Start free, no credit card") + unity ("For founders who move fast").
Referral program: reciprocity (reward both parties) + social proof ("X friends already joined") + unity ("Invite your team") + commitment (ask after a good experience).
See references/case-studies.md for full worked teardowns of multi-principle stacks across industries.
The Ethical Line
Persuasion helps people see value they'd appreciate anyway; manipulation tricks people into choices against their interests. The deciding tests are in the Quick Diagnostic below — run every persuasive element through them before shipping.
See references/ethics.md when a tactic feels borderline or you ship to vulnerable users — the persuasion-vs-manipulation decision tree, the regulatory landscape (FTC, GDPR, DSA, dark-patterns law), vulnerable-population safeguards, and audit templates.
Common Mistakes
| Mistake | Why It Fails | Fix |
|---|---|---|
| Fake social proof | Destroys trust when discovered | Use real data or don't use it |
| Overuse of scarcity | Becomes noise, loses power | Reserve for genuine urgency |
| Inconsistent authority | Undermines credibility | Don't claim expertise you lack |
| Forced reciprocity | Feels transactional, not genuine | Give without immediate ask |
| Generic unity | "Everyone" is not a tribe | Define specific shared identity |
Quick Diagnostic
Audit any persuasive element:
| Question | If No | Action |
|---|---|---|
| Which principle(s) am I using? | You're relying on luck | Explicitly design for influence |
| Am I combining principles? | Missing leverage | Layer multiple principles |
| Is this claim/tactic truthful? | You're manipulating | Remove or replace with truth |
| Does it help the user (not just convert)? | You're exploiting, not persuading | Realign the tactic with the user's goal |
| Would it still work if the user knew the strategy? | The tactic relies on deception | Replace with a transparent version |
| Can users easily reverse the decision? | Ethical concern | Add clear opt-outs |
| Safe for vulnerable users (children, elderly, distressed)? | Heightened-harm risk | Apply ethics.md safeguards or exclude them |
Further Reading
Based on Robert Cialdini's research and books:
- "Influence: The Psychology of Persuasion" by Robert B. Cialdini (Original + Expanded Edition with Unity principle)
- "Pre-Suasion: A Revolutionary Way to Influence and Persuade" by Robert B. Cialdini (Advanced: creating privileged moments for influence)
About the Author
Robert B. Cialdini, PhD is Regents' Professor Emeritus of Psychology and Marketing at Arizona State University. Influence, the foundational text on persuasion science, has sold over 5 million copies worldwide, and he has consulted for Fortune 500 companies, government agencies, and nonprofits on ethical influence.
GitHub 저장소
자주 묻는 질문
influence-psychology Skill이란 무엇인가요?
influence-psychology은(는) wondelai이(가) 만든 Claude Skill입니다. Skill은 Claude가 필요할 때 불러오는 지침과 리소스를 묶어 추가 프롬프트 없이 influence-psychology 관련 작업을 수행할 수 있게 합니다.
influence-psychology은(는) 어떻게 설치하나요?
이 페이지의 설치 명령을 사용하세요. influence-psychology을(를) Claude Code 플러그인으로 추가하거나 저장소를 skills 디렉터리에 복제한 다음 Claude를 다시 시작해 Skill을 불러옵니다.
influence-psychology은(는) 어떤 카테고리에 속하나요?
influence-psychology은(는) 테스팅 카테고리에 속합니다.
influence-psychology은(는) 무료로 사용할 수 있나요?
네. influence-psychology은(는) AIMCP에 등록되어 있으며 무료로 설치할 수 있습니다.
연관 스킬
이 Claude Skill은 MMLU, GSM8K를 포함한 60개 이상의 표준화된 학술 과제에서 LLM 성능을 벤치마크하기 위해 lm-evaluation-harness를 실행합니다. 개발자들이 모델 품질을 비교하고, 학습 진행 상황을 추적하거나 학술 결과를 보고할 수 있도록 설계되었습니다. 이 도구는 HuggingFace와 vLLM 모델을 포함한 다양한 백엔드를 지원합니다.
이 스킬은 cron 표현식을 사용하여 Worker를 스케줄링하기 위한 Cloudflare Cron Triggers 구현에 관한 포괄적인 지식을 제공합니다. 주기적 작업, 유지보수 작업, 자동화된 워크플로우 설정 방법을 다루며, 잘못된 cron 표현식이나 시간대 문제 같은 일반적인 이슈들을 해결하는 방법을 포함합니다. 개발자들은 이를 통해 스케줄된 핸들러 구성, cron 트리거 테스트, Workflows 및 Green Compute와의 연동 작업을 수행할 수 있습니다.
이 Claude Skill은 Python 스크립트를 통해 로컬 웹 애플리케이션을 테스트하기 위한 Playwright 기반 툴킷을 제공합니다. 프론트엔드 검증, UI 디버깅, 스크린샷 캡처, 로그 확인 기능을 지원하며 서버 라이프사이클을 관리합니다. 브라우저 자동화 작업에 사용하되 컨텍스트 오염을 방지하기 위해 소스 코드를 읽지 않고 스크립트를 직접 실행하세요.
이 스킬은 테스트 통과를 확인한 후 체계적인 통합 옵션을 제시하여 개발자가 완성된 작업을 마무리하도록 돕습니다. 구현이 완료된 후 머지, PR 생성, 브랜치 정리와 같은 워크플로우를 안내합니다. 코드가 준비되고 테스트가 완료되었을 때 개발 프로세스를 체계적으로 마무리하기 위해 사용하세요.
