generate-creative-ideas

Compare original and translation side by side

🇺🇸

Original

English

🇨🇳

Translation

Chinese

Creativity Skill

创意技能

Workflow

工作流程

1. Ask       → "What problem/challenge?"
2. Context   → Understand current state BEFORE suggesting
3. Diagnose  → Match situation to technique(s)
4. Generate  → Walk through technique step-by-step
5. Evaluate  → Score and filter ideas
6. Develop   → Shape top ideas into actionable concepts
7. Output    → Structured ideas + next actions (Thai if user uses Thai)

Context questions (step 2):

"What do you currently do?"
"What have you tried?"
"What's working? What's not?"

Anti-pattern: Jumping to solutions without understanding context = generic noise

1. 询问       → "你面临的问题/挑战是什么？"
2. 了解背景   → 在给出建议前先理解当前状况
3. 诊断匹配  → 将当前场景与合适的技术方法匹配
4. 创意生成  → 逐步讲解技术方法的应用步骤
5. 评估筛选  → 为创意打分并筛选优质方案
6. 方案完善  → 将优质创意转化为可落地的具体概念
7. 输出结果  → 结构化的创意+后续行动（若用户使用泰语则输出泰语内容）

背景调研问题（步骤2）：

"你目前正在做什么？"
"你已经尝试过哪些方法？"
"哪些方法有效？哪些无效？"

反模式：不了解背景就直接给出解决方案=泛泛而谈的无用内容

Situation -> Technique Matrix

场景-技术方法匹配矩阵

Situation	Techniques	Combination Recipe
Stuck / No ideas	Random Word, Forced Connections, Oblique Strategies	Random Word -> Forced Connections -> Dot Vote
Need breakthrough	First Principles, Challenging Assumptions, Combinatorial Engine	HMW -> Worst Idea -> SCAMPER -> Brainwriting
Improve existing	SCAMPER, Reverse Brainstorming, TRIZ-AI	SCAMPER -> Reverse Brainstorm -> Impact/Effort
Explore systematically	Morphological Box, Six Thinking Hats, Constraint Injection	Morphological Box -> Constraint Injection -> Clustering
Reframe problem	How Might We (HMW), Jobs to be Done	HMW -> JTBD -> First Principles
Team ideation	6-3-5 Brainwriting, Multi-Persona Parallel	Mind Map -> Brainwriting -> Affinity -> Multi-Vote
Too many ideas	Impact/Effort Matrix, Idea Clustering, NAF Scoring	Clustering -> NAF Quick Score -> Impact/Effort
AI ideas too similar	Divergence Guard, Constraint Injection, Incubation Cycling	Divergence Guard -> Domain Shift -> Incubation
Technical/engineering	TRIZ-AI, First Principles	TRIZ -> First Principles -> Assumption Mapping
Cross-domain innovation	Combinatorial Engine, Analogical Thinking	Combinatorial Engine -> Analogical -> Constraint
Content ideas (blog/video/course)	Content Pillars, Audience Pain Points, Gap Analysis	Pillars -> Pain Points -> SCAMPER -> Validate
Business/product ideas	JTBD, Opportunity Canvas, Lean Validation	JTBD -> HMW -> Morphological -> ICE Score

场景	适用技术	组合方案
陷入瓶颈/毫无头绪	Random Word, Forced Connections, Oblique Strategies	Random Word -> Forced Connections -> Dot Vote
需要突破性创意	First Principles, Challenging Assumptions, Combinatorial Engine	HMW -> Worst Idea -> SCAMPER -> Brainwriting
优化现有方案	SCAMPER, Reverse Brainstorming, TRIZ-AI	SCAMPER -> Reverse Brainstorm -> Impact/Effort
系统性探索创意	Morphological Box, Six Thinking Hats, Constraint Injection	Morphological Box -> Constraint Injection -> Clustering
重构问题视角	How Might We (HMW), Jobs to be Done	HMW -> JTBD -> First Principles
团队头脑风暴	6-3-5 Brainwriting, Multi-Persona Parallel	Mind Map -> Brainwriting -> Affinity -> Multi-Vote
创意过多难以筛选	Impact/Effort Matrix, Idea Clustering, NAF Scoring	Clustering -> NAF Quick Score -> Impact/Effort
AI生成创意同质化严重	Divergence Guard, Constraint Injection, Incubation Cycling	Divergence Guard -> Domain Shift -> Incubation
技术/工程领域创意	TRIZ-AI, First Principles	TRIZ -> First Principles -> Assumption Mapping
跨领域创新	Combinatorial Engine, Analogical Thinking	Combinatorial Engine -> Analogical -> Constraint
内容创意（博客/视频/课程）	Content Pillars, Audience Pain Points, Gap Analysis	Pillars -> Pain Points -> SCAMPER -> Validate
商业/产品创意	JTBD, Opportunity Canvas, Lean Validation	JTBD -> HMW -> Morphological -> ICE Score

Technique Quick Reference

技术方法速查

Divergent (Generate)

发散型（创意生成）

Technique	One-liner
SCAMPER	7 lenses: Substitute, Combine, Adapt, Modify, Put to other use, Eliminate, Reverse
Random Word	Random noun -> list attributes -> force connections to problem
Reverse Brainstorm	"How to make it worse?" -> Invert each idea
First Principles	Strip to fundamentals -> Rebuild from scratch
Six Hats	6 perspectives: Facts, Feelings, Risks, Benefits, Ideas, Process
HMW	Reframe as "How Might We [verb] for [user] so that [outcome]?"
Morphological Box	Parameters x Variations matrix -> Combine systematically
Analogies	"How does [other domain] solve this?"
Assumptions	List assumptions -> Challenge/invert each
Forced Connections	Combine 2 unrelated concepts
Jobs to be Done	"When [situation], I want [motivation], so I can [outcome]"
Oblique Strategies	Random creative prompts to break deadlocks

技术方法	一句话说明
SCAMPER	7种思考视角：替代、组合、适配、修改、复用、删除、反转
Random Word	随机选取名词→列出其属性→强制关联到当前问题
Reverse Brainstorm	先思考“如何让问题变得更糟？”→再将每个想法反转得到解决方案
First Principles	剥离表象回归本质→从头构建解决方案
Six Hats	6种思考视角：事实、情感、风险、收益、创意、流程
HMW	将问题重构为“我们如何为[用户]实现[结果]？”的形式
Morphological Box	构建参数-变体矩阵→系统性组合生成创意
Analogies	思考“其他领域是如何解决类似问题的？”
Assumptions	列出当前的所有假设→逐一质疑或反转
Forced Connections	将两个不相关的概念进行组合
Jobs to be Done	用“当[场景]时，我想要[动机]，从而实现[结果]”梳理需求
Oblique Strategies	用随机创意提示打破思维僵局

AI-Optimized (Generate + Diversify)

AI优化型（创意生成+多样化）

Technique	One-liner
Incubation Cycling	Generate -> Pause -> Fresh restart (no prior context) -> Compare
Combinatorial Engine	Abstract -> Retrieve 3 domains -> Generalize -> Combine -> Instantiate
Multi-Persona Parallel	Run 4+ personas SIMULTANEOUSLY (not sequentially)
Constraint Injection	Add random constraint -> Force novel solutions
Divergence Guard	Force opposite -> Domain shift -> Absurdity injection
TRIZ-AI	Apply inventive principles (segmentation, nesting, dynamization, etc.)
Tree of Thoughts	Explore multiple reasoning branches simultaneously (74% vs CoT 49%)

技术方法	一句话说明
Incubation Cycling	生成创意→暂停→重启（不参考之前的内容）→对比优化
Combinatorial Engine	抽象问题→检索3个相关领域→归纳共性→组合创新→落地细化
Multi-Persona Parallel	同时模拟4个以上用户角色进行创意（而非依次模拟）
Constraint Injection	添加随机限制条件→倒逼产生新颖解决方案
Divergence Guard	强制生成反向创意→切换领域→注入荒诞元素打破同质化
TRIZ-AI	应用发明原理（如分割、嵌套、动态化等）
Tree of Thoughts	同时探索多条推理分支（创意质量优于CoT，提升至74% vs 49%）

Convergent (Evaluate + Refine)

收敛型（评估+完善）

Technique	One-liner
NAF Quick Score	Rate Novelty + Attractiveness + Feasibility (1-10 each)
Impact/Effort Matrix	2x2: Quick Wins, Big Bets, Fill-ins, Avoid
ICE Scoring	Impact x Confidence x Ease (1-10 each)
Idea Clustering	Group similar -> Name clusters -> Pick best from each
Dot Voting	Each person gets 3-5 votes -> Surface favorites
Assumption Mapping	Map assumptions on Importance x Certainty -> Test riskiest first

技术方法	一句话说明
NAF Quick Score	从新颖性、吸引力、可行性三个维度评分（每项1-10分）
Impact/Effort Matrix	2x2矩阵分类：快速见效、重大投入、补充优化、避免尝试
ICE Scoring	从影响力、信心、易用性三个维度评分（每项1-10分）并相乘
Idea Clustering	将相似创意分组→为每组命名→从每组中选取最优创意
Dot Voting	每人分配3-5票→选出最受欢迎的创意
Assumption Mapping	将假设映射到重要性-确定性矩阵→优先验证高重要性低确定性的假设

Idea Evaluation

创意评估

After generating ideas, ALWAYS offer evaluation. Default to NAF Quick Score.

生成创意后，必须进行评估。默认使用NAF快速评分法。

NAF Quick Score (Default)

NAF快速评分法（默认）

Criterion	Question	Scale
Novelty	How new/surprising is this?	1-10
Attractiveness	How well does it solve the problem?	1-10
Feasibility	How realistic to implement?	1-10

Interpretation:

Total 24-30: Strong candidate -> develop further
Total 18-23: Promising -> refine or combine
Total < 18: Weak -> park or discard
Feasibility 8+: Worth trying (remaining 20% is implementation)
High N+A but low F: Reframe feasibility barriers as new problems to solve

评估维度	对应问题	评分范围
新颖性	这个创意的新颖/惊喜程度如何？	1-10
吸引力	这个创意能多大程度解决问题？	1-10
可行性	这个创意的落地难度如何？	1-10

评分解读：

总分24-30：优质候选方案→进一步完善
总分18-23：有潜力的方案→优化或与其他创意组合
总分<18：薄弱方案→暂存或舍弃
可行性得分8+：值得尝试（剩余20%为落地执行问题）
新颖性+吸引力得分高但可行性低：将可行性障碍重构为新问题解决

When to Use Which Evaluation

不同场景的评估方法选择

Situation	Method
Quick screening (5+ ideas)	NAF or Dot Voting
Growth experiments	ICE Scoring
Data-driven product decisions	RICE Scoring
Complex multi-criteria	Weighted Scoring Matrix
Visual team alignment	Impact/Effort 2x2

Rules:

NEVER evaluate during divergent phase -- generate first, judge later
Take a break between generating and evaluating (different mindset)
Different idea types need different criteria (incremental vs disruptive)

Details: evaluation.md

场景	适用方法
快速筛选（5个以上创意）	NAF评分或Dot Voting
增长实验	ICE评分法
数据驱动的产品决策	RICE评分法
复杂多维度评估	加权评分矩阵
团队可视化对齐	Impact/Effort 2x2矩阵

规则：

发散阶段绝对不要评估→先生成所有创意，再进行评判
创意生成和评估之间要间隔一段时间（切换思维模式）
不同类型的创意需要不同的评估标准（渐进式vs颠覆性）

详情：evaluation.md

Content Creator Mode

内容创作者模式

When ideating for blog posts, videos, courses, or social media:

当为博客、视频、课程或社交媒体构思创意时：

Step 1: Define Content Pillars (3-5 themes)

步骤1：定义内容支柱（3-5个核心主题）

Pillar = Core theme that reflects expertise + audience needs
Example: Excel -> [Formulas, Data Viz, Automation, Tips & Tricks, Career]

内容支柱 = 体现专业能力+满足受众需求的核心主题
示例：Excel领域→[公式、数据可视化、自动化、技巧与窍门、职业发展]

Step 2: Audience-First Ideation

步骤2：以受众为中心的创意构思

Source	Questions
Pain Points	What frustrates them most?
Questions	What do they repeatedly ask?
Gaps	What's poorly explained by competitors?
Wishes	What do they wish existed?
Mistakes	What common errors do they make?

信息来源	调研问题
痛点	受众最困扰的问题是什么？
提问	受众反复询问的问题是什么？
空白	竞争对手讲解不到位的内容是什么？
期望	受众希望存在的内容是什么？
错误	受众常犯的错误是什么？

Step 3: Generate Ideas (use any technique from matrix)

步骤3：生成创意（使用矩阵中的任意技术方法）

Apply creativity techniques TO the content pillars:

SCAMPER on existing popular content
Reverse brainstorm: "How to make the worst tutorial?"
Analogies: "How would Netflix teach Excel?"

将创意技术应用到内容支柱上：

对现有热门内容使用SCAMPER方法
反向头脑风暴：“如何制作最糟糕的教程？”
类比思考：“Netflix会如何教授Excel？”

Step 4: Validate Before Creating

步骤4：创作前验证

[ ] Search demand? (keyword research)
[ ] Real audience pain point? (comments, surveys)
[ ] Can I add unique value? (gap analysis)
[ ] Fits my pillars? (strategy alignment)
[ ] Would I click this? (title/thumbnail test)
[ ] Can be repurposed? (1 piece -> 5+ formats)

[ ] 有搜索需求吗？（关键词调研）
[ ] 是真实的受众痛点吗？（评论、调研）
[ ] 我能提供独特价值吗？（空白分析）
[ ] 符合我的内容支柱吗？（战略对齐）
[ ] 我会点击这个内容吗？（标题/缩略图测试）
[ ] 可复用吗？（1份内容转化为5+种格式）

Repurposing Chain

内容复用链条

Blog post -> YouTube video -> Shorts/Reels -> Social posts -> Email -> Course module

Details: content-ideation.md

博客文章 → YouTube视频 → Shorts/Reels短视频 → 社交平台帖子 → 邮件推送 → 课程模块

详情：content-ideation.md

Idea Development Pipeline

创意完善流程

Shape raw ideas into actionable concepts:

[Generate] -> [Cluster] -> [Evaluate] -> [Develop] -> [Validate] -> [Execute]

将原始创意转化为可落地的方案：

[生成创意] -> [分组聚类] -> [评估筛选] -> [方案完善] -> [验证测试] -> [执行落地]

Quick Concept Card (for top ideas)

快速创意卡片（针对优质创意）

IDEA: [Name]
TAGLINE: [One compelling sentence]
PROBLEM: [What pain it solves]
SOLUTION: [How it works - 2-3 sentences]
TARGET USER: [Who benefits]
KEY INSIGHT: [The "aha" behind this]
EFFORT: [S / M / L]
BIGGEST RISK: [What could go wrong]
QUICKEST TEST: [How to validate cheaply]
NEXT STEP: [One concrete action]

创意名称: [名称]
宣传语:  [一句有吸引力的话]
解决问题: [解决的核心痛点]
解决方案: [运作方式 - 2-3句话]
目标用户: [受益人群]
核心洞察: [创意背后的“顿悟点”]
落地难度: [小 / 中 / 大]
最大风险: [可能出现的问题]
快速测试: [低成本验证的方法]
下一步行动: [具体的落地动作]

Assumption Mapping (for important ideas)

假设映射（针对重要创意）

List assumptions: "What must be true for this to work?"
Categorize: Desirability / Feasibility / Viability
Map on 2x2: Importance (high/low) x Certainty (high/low)
Test high-importance + low-certainty FIRST

Details: idea-pipeline.md

列出假设：“要让这个创意成功，哪些条件必须成立？”
分类：吸引力 / 可行性 / 商业可行性
映射到2x2矩阵：重要性（高/低）×确定性（高/低）
优先验证高重要性+低确定性的假设

详情：idea-pipeline.md

Output Formats

输出格式

Use these when presenting ideas to the user:

向用户展示创意时使用以下格式：

Quick List (5+ ideas)

快速列表（5个以上创意）

1. **[Idea Name]** -- [One-line description]
2. **[Idea Name]** -- [One-line description]
...

1. **[创意名称]** -- [一句话描述]
2. **[创意名称]** -- [一句话描述]
...

Scored List (after evaluation)

评分列表（评估后）

| # | Idea | N | A | F | Total |
|---|------|---|---|---|-------|
| 1 | ...  | 8 | 9 | 7 | 24    |

| 序号 | 创意 | 新颖性 | 吸引力 | 可行性 | 总分 |
|---|------|---|---|---|-------|
| 1 | ...  | 8 | 9 | 7 | 24    |

Concept Cards (top 3 ideas)

创意卡片（Top3创意）

Use the Quick Concept Card format above for each top idea.

对每个优质创意使用上述快速创意卡片格式。

Comparison Matrix (deciding between options)

对比矩阵（方案抉择时）

| Criteria     | Idea A | Idea B | Idea C |
|-------------|--------|--------|--------|
| Novelty     | 4/5    | 3/5    | 5/5    |
| Feasibility | 3/5    | 5/5    | 2/5    |
| Impact      | 5/5    | 3/5    | 4/5    |

| 评估维度     | 创意A | 创意B | 创意C |
|-------------|--------|--------|--------|
| 新颖性     | 4/5    | 3/5    | 5/5    |
| 可行性 | 3/5    | 5/5    | 2/5    |
| 影响力      | 5/5    | 3/5    | 4/5    |

AI Creativity Guidelines

AI创意应用指南

Key Research Insights (2025-2026)

核心研究发现（2025-2026）

Finding	Implication
Ask AI HOW to think, not WHAT to think	Process prompts > product prompts
Human-first ideation preserves diversity	User brainstorms first, THEN AI expands
LLMs match average creativity (52nd percentile)	AI is a partner, not a replacement
Homogenization effect (g ~ -0.86)	Use Divergence Guard actively
"Creative Scar" -- creativity drops after AI withdrawal	Don't outsource all creative thinking
Tree of Thoughts: 74% vs CoT 49%	Use ToT for complex creative tasks
Multi-LLM collaboration enhances originality	Stack techniques, vary approaches
Only 0.28% of LLM ideas reach top 10% human creativity	Push past first outputs aggressively

研究结论	实践启示
告诉AI如何思考而非思考什么	过程型提示词优于结果型提示词
以人类为主的构思能保留创意多样性	先由人类头脑风暴，再用AI拓展
大语言模型的创意水平相当于人类平均水平（第52百分位）	AI是合作伙伴，而非替代者
同质化效应（相关系数g≈-0.86）	主动使用Divergence Guard避免同质化
“创意断层”——过度依赖AI后人类创意能力下降	不要外包所有创意思考工作
Tree of Thoughts方法：创意质量提升至74%（对比CoT的49%）	复杂创意任务使用ToT方法
多LLM协作能提升创意原创性	组合使用多种技术方法，变换思路
仅0.28%的LLM生成创意能达到人类前10%的创意水平	要主动突破AI的初始输出内容

AI Role Rules

AI角色规则

Best as: Exploration partner, constraint enforcer, analogy finder, elaborator
Avoid: Sole ideation source, final decision maker
Timing: After initial human ideation, not before
Override rate: Maintain 15-25% human override for optimal outcomes

最佳定位： 探索伙伴、限制条件施加者、类比发现者、创意细化者
避免： 唯一创意来源、最终决策者
使用时机： 人类完成初始构思后，而非之前
人工干预比例： 保持15-25%的人工干预以获得最佳结果

Anti-Patterns

反模式

Don't	Do Instead
"Give me ideas for X"	"Give me 10 ideas for X that would surprise an expert"
Accept first outputs	Push past 2-3 rounds; first ideas are "greatest hits"
Use AI before thinking	Brainstorm independently first, then expand with AI
"Be creative"	Use specific technique (SCAMPER, constraints, personas)

Details: methodology.md | prompt-templates.md

错误做法	正确做法
“给我一些关于X的创意”	“给我10个能让专家感到惊喜的X领域创意”
接受AI的初始输出	至少推进2-3轮；初始创意都是“热门内容复刻”
先用AI再自己思考	先独立头脑风暴，再用AI拓展创意
“要有创意”	使用具体技术方法（SCAMPER、限制条件、用户角色等）

详情：methodology.md | prompt-templates.md

References

参考资料

Step-by-step guides: techniques.md
AI methodology & research: methodology.md
Random word bank: random-words.md
Evaluation frameworks: evaluation.md
Content creator ideation: content-ideation.md
Idea development pipeline: idea-pipeline.md
AI prompt templates: prompt-templates.md

分步指南: techniques.md
AI方法论与研究: methodology.md
随机词汇库: random-words.md
评估框架: evaluation.md
内容创作者创意: content-ideation.md
创意完善流程: idea-pipeline.md
AI提示词模板: prompt-templates.md