Zhihu Hot Article Engine
Write Zhihu articles that rank on the hot list and earn genuine upvotes.
When to use
- Writing a comprehensive Zhihu answer that needs to stand out among hundreds
- Creating a long-form Zhihu article (专栏文章) targeting a trending topic
- Analyzing a Zhihu question to determine whether it's worth answering
- Optimizing an existing answer that isn't getting traction
- Writing a hot-list strategy: timing, keyword density, and engagement baiting (legitimate)
Core Principles
1. Zhihu Writing is Different
Zhihu is NOT a blog, NOT a social media post. It's a credibility-first knowledge platform:
| Aspect | Blog | Zhihu | |
|---|---|---|---|
| Reader mindset | Skimming for info | Scanning for entertainment | Seeking authoritative, structured answers |
| Required depth | Medium | Very low | Very high |
| Personal story | Optional | Common | Structure anchor |
| Data/citations | Nice to have | Never | Essential |
| Update cycle | Once | Once | Living document (edit/sticky) |
| Length sweet spot | 1500-2500 words | 280 chars | 3000-8000+ words |
| Credibility signal | Domain | Followers | Credentials + data + experience |
2. The Zhihu Algorithm
Zhihu's recommendation system ranks answers based on these factors (ordered by importance):
- Upvote count + velocity — fast early upvotes = massive amplification
- Answer length — longer answers (>3000 chars) consistently rank better
- Creator credibility — verified credentials, past high-quality answers
- Dwell time — how long readers stay on your answer (scroll depth)
- Engagement — comments, collections, shares (collections = strong signal)
- Relevance — how well your answer matches the question's topic cluster
- Freshness — new answers get a temporary boost for first 48 hours
Key insight: The first 2 hours determine whether your answer goes viral or dies.
3. The Zhihu Format (Structural Canon)
Every great Zhihu answer follows this structural pattern:
╔══════════════════════════════════════╗
║ HEADER ║
║ ┌──────────────────────────────────┐║
║ │ 谢邀 / 谢不邀 │║ ← Credibility signal
║ │ [Credentials: 10年从业经验/PhD] │║ ← Authority anchor
║ └──────────────────────────────────┘║
╠══════════════════════════════════════╣
║ HOOK (First 3 lines) ║
║ ┌──────────────────────────────────┐║
║ │ 直接回答:「答案是:……」 │║ ← Direct answer first
║ │ or │║
║ │ 先讲结论:「先说结论:是的。」 │║ ← Thesis statement
║ └──────────────────────────────────┘║
╠══════════════════════════════════════╣
║ STORY / Personal Anchor ║
║ ┌──────────────────────────────────┐║
║ │ "我是怎么知道这个的" │║ ← Personal credibility
║ │ [Concrete personal experience] │║ ← Lived example
║ └──────────────────────────────────┘║
╠══════════════════════════════════════╣
║ STRUCTURED ARGUMENT ║
║ ┌──────────────────────────────────┐║
║ │ 1. 核心原理 │║ ← First principle
║ │ [Data/citation + explanation] │║
║ │ 2. 常见误区 │║ ← Refute against false theses
║ │ [Counter-argument + why wrong] │║
║ │ 3. 实用建议 │║ ← Actionable framework
║ │ [Step-by-step methodology] │║
║ │ 4. 深层思考 │║ ← Meta-analysis (the X factor)
║ │ [Unique insight] │║
║ └──────────────────────────────────┘║
╠══════════════════════════════════════╣
║ CLOSURE ║
║ ┌──────────────────────────────────┐║
║ │ 总结一句话:「……。」 │║ ← One-line summary
║ │ 延伸阅读:[book/paper/link] │║ ← Bonus value
║ │ "如果觉得有用,点个赞/关注" │║ ← Soft CTA
║ └──────────────────────────────────┘║
╚══════════════════════════════════════╝
Phase 0: Knowledge Reconnaissance
BEFORE writing, you MUST complete this research phase. Each item is mandatory.
Step 1: Question Analysis
Scan the question:
- Is it a hot question? (check Zhihu hot list relevance)
- Is it well-formulated? (clear, specific, debatable)
- What's the question type?
- 事实类 ("What is X?" ) → data-heavy, authoritative sources
- 观点类 ("Do you think X?") → personal experience + reasoning
- 经验类 ("How to do X?") → step-by-step, credibility via results
- 评价类 ("How do you evaluate X?") → balanced, multi-perspective
- Who are the top answerers? (read 3-5 top answers to understand the landscape)
Step 2: Knowledge Mapping
Search for:
- Core concepts: definitions, principles, academic references
- Data points: statistics, studies, surveys with specific numbers
- Case studies: real-world examples, historical analogues
- Expert opinions: what credible figures have said on this topic
Step 3: Counter-Intuition Hunt
MUST find at least one counter-intuitive point:
- "Commonly believed X, but actually Y"
- "Most people think A, but research shows B"
- "The mainstream advice is wrong because..."
Step 4: Citation Harvesting
Collect minimum 3 high-credibility citations:
- Academic papers (preferred)
- Official data/statistics
- Authoritative media reports
- Expert quotes with named attribution
Step 5: Angle Selection
Choose your UNIQUE angle:
- Contrarian angle: Challenge the mainstream view
- Expert insider: Reveal what people don't know from the inside
- Systematizer: Provide a framework/step-by-step no one else has
- Deep-diver: Go 10x deeper than existing answers
- Connector: Connect two seemingly unrelated fields
Phase 1: Structure Building
After Phase 0, build the answer structure:
## Structure Template
### Opening (150-250 words)
【谢邀/直接回答】
[Thesis: one sentence that answers the question directly]
[Credentials: why YOU should be trusted on this]
[Map: "我将从三个方面回答这个问题:……"]
### Body Section 1: Core Insight (400-800 words)
[概念定义 + 原理说明]
[数据/研究支持]
[个人经历或案例印证]
【要点总结:一个小绿框/加粗句】
### Body Section 2: Counter Arguments & Refinements (300-600 words)
[常见错误认知]
[为什么这些是错的]
[正确理解应该是什么]
### Body Section 3: Practical Framework (400-800 words)
[可操作的步骤/方法]
[每个步骤的具体应用场景]
[注意事项和边界条件]
### Meta-Angle: The Deep Insight (200-500 words)
[这个问题的本质是什么?]
[大多数回答没有触及到的层面]
[——这是让你的回答被收藏的段落]
### Closing (100-200 words)
[一句话总结]
[延伸阅读推荐]
[互动的邀请]
Phase 2: Writing Execution
Writing Rules for Zhihu
| Rule | Explanation |
|---|---|
| Direct answer first | Never bury the lead. First 3 lines must answer the question |
| 分段≤5句 | Short paragraphs for mobile readability |
| 加粗关键句 | Bold the thesis statement in each section |
| 数据必须有来源 | Every number needs a citation |
| 每300字一个"钩子" | A surprising fact, a question, or a provocative statement |
| 专业但不学院 | Use plain language, but show depth |
| 例子≥1个 | Every abstract point needs a concrete example |
| 字数3000+ | Short answers don't rank. Aim for 3000-8000 chars |
| 不要纯讲道理 | Mix personal story with data |
| 留白 | Not every point needs exhaustive explanation. Let readers fill gaps |
Title Rules
For 专栏文章 (not answers), the title is critical:
- 含关键词: Include the topic's core search keyword
- 数字 + 结果: "3个方法","10年经验","5步做到"
- 悬念/冲突: "为什么专业的人都不说真话?"
- 具体而非抽象: "我如何从零学会XX" vs "学习XX的方法"
- 情绪触发: 好奇、紧迫、认同、优越
- A/B测试: Write 3-5 title variants and pick the best
Engagement Optimization
- Asking questions: End sections with a rhetorical question to boost comment engagement
- Comments bait: "有不同意见欢迎评论区讨论"
- Collection candy: "建议先收藏,慢慢看" (at a logical mid-point)
- Upvote ask: Subtle, at the end, with justification ("如果对你有帮助")
Quality Checklist
- Phase 0 completed (question analysis, knowledge mapping, citations found)
- Counter-intuitive point identified and woven in
- Direct answer in first 3 lines
- Minimum 3 high-credibility citations
- Personal story/experience anchor
- At least 1 concrete example per abstract point
- 3000+ characters length confirmed
- Short paragraphs (≤5 sentences each)
- Key clauses bolded for mobile scanning
- Every number has a source
- Title optimized (for 专栏) or opening optimized (for 回答)
- Comment/engagement hooks placed strategically
- Soft CTA at the end
- No self-promotion unless explicitly relevant
- Read aloud test: sounds natural, not translated from English
Advanced Tips
Hot-List Timing
- Post timing: Weekday evenings 19:00-22:00 (maximum active users)
- Weekend: Saturday 10:00-14:00 (information-seeking mode)
- Avoid: Late night (00:00-06:00) and working hours (09:00-12:00)
- Boost window: First 2 hours after posting — engage with every comment
Building Credibility on a New Account
- Answer 3-5 smaller questions FIRST to build answer history
- Get your first 100 upvotes on easy questions before tackling big ones
- Credentials: Verify your real identity (知乎认证) if applicable
- Consistency: Answer 1-2 questions per week minimum
Handling Controversial Topics
- Acknowledge opposing views: "很多人可能觉得……"
- Present data, not opinion: "根据研究显示……"
- Stay calm: never use aggressive language
- "我可能不对,但目前的理解是……"
- Invite counter-arguments: "欢迎补充或反驳"
