yueqian/wiki
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

init: 导入团队知识库内容

Browse Source
This commit is contained in:
yueqian-ai
2026-05-14 16:56:48 +08:00
commit acca2041f0
1681 changed files with 285734 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files
.claude/skills/deep-research/research-notes/registry-c.md
+44
View File
@@ -0,0 +1,44 @@
# CITATION REGISTRY
Built from: task-c.md
## Approved Sources
[1] 拼多多社交电商商业模式深度解析: 拼团裂变、下沉市场与算法推荐机制 — Sina财经 | https://cj.sina.cn/articles/view/7879848900/1d5acf3c401902we8s | Source-Type: secondary-industry | As Of: 2024-06 | Auth: 6 | From: task-c
[2] 实战: 解析互联网拼团裂变模式 — 知乎专栏 | https://zhuanlan.zhihu.com/p/258546036 | Source-Type: community | As Of: 2020-09 | Auth: 6 | From: task-c
[3] "社交+电商"模式的裂变与挑战——以拼多多为例 — 汉斯出版社(学术) | https://www.hanspub.org/journal/paperinformation?paperID=130599 | Source-Type: academic | As Of: 2024 | Auth: 7 | From: task-c
[4] 拼多多屡试不爽的裂变，你为什么玩不好？ — 顶象 | https://www.dingxiang-inc.com/blog/post/297 | Source-Type: secondary-industry | As Of: 2023 | Auth: 7 | From: task-c
[5] 拼团、砍价、分销｜用户要增长，裂变和转化才是关键 — 人人都是产品经理 | https://www.woshipm.com/operate/2295477.html | Source-Type: community | As Of: 2023 | Auth: 6 | From: task-c
[6] 社交裂变玩法中拼团与分销功能设计的优秀案例分享 — 搜狐 | https://www.sohu.com/a/902973212_100150288 | Source-Type: journalism | As Of: 2025-04 | Auth: 5 | From: task-c
[7] 拼多多在用户裂变增长模式上的独特方式方法及案例分析 — 商派 | https://www.shopex.cn/news/archives/17226.html | Source-Type: secondary-industry | As Of: 2023 | Auth: 6 | From: task-c
[8] 什么是代言人与合伙人模式？微三云深度解析 — 腾讯云 | https://cloud.tencent.com/developer/news/1014119 | Source-Type: secondary-industry | As Of: 2023-04 | Auth: 6 | From: task-c
[9] 2023新消费模式商城"代言人"+"合伙人"模式 — 知乎 | https://zhuanlan.zhihu.com/p/610766506 | Source-Type: community | As Of: 2023-03 | Auth: 5 | From: task-c
[10] 揭秘"排队免单"：消费变投资 — 知乎 | https://zhuanlan.zhihu.com/p/1969786292100370904 | Source-Type: community | As Of: 2024 | Auth: 6 | From: task-c
[11] 排队免单模式全产业适配解析与真实案例复盘 — 搜狐 | https://www.sohu.com/a/1013343290_122699958 | Source-Type: journalism | As Of: 2025 | Auth: 5 | From: task-c
[12] "消费全返"为何存在刑事风险 — 最高人民检察院 | https://www.spp.gov.cn/spp/llyj/201907/t20190717_443375.shtml | Source-Type: official | As Of: 2019-07 | Auth: 9 | From: task-c
[13] 电商微商、消费返利？小心！有可能是新型传销 — 北京科技报 | https://www.bbtnews.com.cn/2021/1220/423361.shtml | Source-Type: journalism | As Of: 2021-12 | Auth: 7 | From: task-c
[14] 国家关于消费返利违法的认定与合规方案 — 启博 | https://www.qibo.com/article/44198.html | Source-Type: secondary-industry | As Of: 2024 | Auth: 6 | From: task-c
[15] 警惕"消费返利"非法集资陷阱 — 北京市委金融委员会办公室 | https://jrj.beijing.gov.cn/jrxcjy/jraq/ffffjzxkt/202405/t20240527_3695341.html | Source-Type: official | As Of: 2024-05 | Auth: 9 | From: task-c
[16] 关于"高额消费返利"类网站涉嫌违法犯罪预警 — 公安部 | https://www.mps.gov.cn:8090/n2253534/n2253535/c4143338/content.html | Source-Type: official | As Of: 2022 | Auth: 10 | From: task-c
[17] 销售模式介绍及法律风险探讨 — 中伦律师事务所 | https://www.zhonglun.com/research/articles/9588.html | Source-Type: secondary-industry | As Of: 2024 | Auth: 8 | From: task-c
[18] 人民法院依法惩治金融犯罪典型案例 — 最高人民法院 | https://www.court.gov.cn/zixun/xiangqing/372731.html | Source-Type: official | As Of: 2023 | Auth: 9 | From: task-c
[19] What's behind the success of China's social commerce app Pinduoduo — Econsultancy | https://econsultancy.com/what-s-behind-the-success-of-china-s-social-commerce-app-pinduoduo/ | Source-Type: secondary-industry | As Of: 2020-05 | Auth: 7 | From: task-c
[20] Pinduoduo and the rise of social e-commerce — YC Startup Library | https://www.ycombinator.com/library/2z-pinduoduo-and-the-rise-of-social-e-commerce | Source-Type: secondary-industry | As Of: 2020 | Auth: 7 | From: task-c
[21] 推三返一模式详细剖析 — 知乎 | https://zhuanlan.zhihu.com/p/1889004163456806973 | Source-Type: community | As Of: 2024 | Auth: 5 | From: task-c
[22] Fun消费返利模式：误区解析与实战策略 — 启博 | https://www.qibo.com/article/44239.html | Source-Type: secondary-industry | As Of: 2024 | Auth: 6 | From: task-c
[23] 城市酷选排队免单 — 知乎 | https://zhuanlan.zhihu.com/p/13117000993 | Source-Type: community | As Of: 2024 | Auth: 5 | From: task-c
[24] 教你读懂代言人＋合伙人模式 — 腾讯云 | https://cloud.tencent.com/developer/news/1073579 | Source-Type: secondary-industry | As Of: 2023 | Auth: 6 | From: task-c
## Dropped
No sources dropped. All 24 sources meet lightweight threshold (Auth >= 5, diverse domains).
## Stats
Total evaluated: 24
Approved: 24
Dropped: 0
Unique domains: 13
Source-type: official 4 / academic 1 / secondary-industry 11 / journalism 3 / community 5
Max single-source share: 5/24 = 21% (pass - under 30% lightweight limit)
Official share: 4/24 = 17% — slightly below 20% lightweight target but acceptable given the topic (regulatory/legal content from government sources is well-represented)
.claude/skills/deep-research/research-notes/task-a.md
+174
View File
@@ -0,0 +1,174 @@
# Research Notes: Chinese Social Commerce Distribution Models — Network Structure Analysis
**Date:** 2026-05-06
**Task:** Research 二加一连锁模式, 双轨制, 太阳线, 矩阵制 — mechanics, profit-sharing, use cases, pros/cons, compliance risks
---
## Sources
| # | URL | Source-Type | Accessibility | Date | Authority (1-10) |
|---|-----|-------------|---------------|------|-------------------|
| 1 | https://zhuanlan.zhihu.com/p/679418297 | Zhihu article (industry analysis) | Open | 2023 | 6 |
| 2 | https://zhuanlan.zhihu.com/p/572293043 | Zhihu article (legal analysis) | Open | 2023 | 6 |
| 3 | https://zhuanlan.zhihu.com/p/691665089 | Zhihu article (compliance Q&A) | Open | 2023 | 6 |
| 4 | https://www.weifenxiao.com/info/display/1954.html | Vendor site (Qibo/Weifenxiao) | Open | 2023-2024 | 5 |
| 5 | https://www.sohu.com/a/609318392_121606784 | Sohu article (legal analysis) | Open | 2023 | 5 |
| 6 | https://www.jubaozan.com/CaseDismantling/info.aspx?itemid=1065 | Vendor analysis page | Open | 2023 | 5 |
| 7 | https://developer.aliyun.com/article/1633401 | Aliyun developer article | Open | 2024-2025 | 6 |
| 8 | https://www.crmeb.com/ask/thread/40153 | CRMEB forum thread | Open | 2023 | 5 |
| 9 | https://cloud.tencent.com/developer/article/2577487 | Tencent Cloud developer article | Open | 2024-2025 | 6 |
| 10 | https://www.qibo.com/article/44931.html | Qibo vendor encyclopedia | Open | 2024 | 5 |
| 11 | https://www.sohu.com/a/277371611_100299982 | Sohu article (all MLM plans) | Open | 2018 | 5 |
| 12 | https://www.163.com/dy/article/G9GDCGLJ05428T8W.html | NetEase article (all plans) | Open | 2023 | 5 |
| 13 | https://wiki.mbalib.com/zh-tw/%E5%8F%8C%E8%BD%A8%E5%88%B6%E5%BA%A6 | MBA智库百科 (encyclopedia) | Open | Ongoing | 7 |
| 14 | https://zhuanlan.zhihu.com/p/554789751 | Zhihu article (微三云 analysis) | Open | 2022 | 5 |
| 15 | https://blog.csdn.net/lrj20030206/article/details/130014508 | CSDN technical blog | Open | 2023 | 5 |
| 16 | https://xueqiu.com/9448294601/238654967 | Xueqiu (financial analysis) | Open | 2023 | 6 |
| 17 | https://www.100ec.cn/detail--6638350.html | 网经社 legal analysis | Open | 2020 | 7 |
| 18 | https://www.lexology.com/library/detail.aspx?g=6f2dfc03-c164-47d8-8571-3876489b583c | Lexology (law firm) | Open | 2024 | 8 |
| 19 | https://www.dehenglaw.com/CN/tansiaocontent/0008/030694/7.aspx | DeHeng law firm analysis | Open | 2020 | 8 |
| 20 | https://maimai.cn/article/detail?fid=1853704803 | MaiMai (professional network) | Open | 2023 | 4 |
| 21 | https://m.92hi.com/news/7577.html | 商城网站建设 (vendor blog) | Open | 2024 | 4 |
| 22 | https://m.92hi.com/news/7578.html | 商城网站建设 (vendor blog) | Open | 2024 | 4 |
| 23 | https://zhuanlan.zhihu.com/p/561439553 | Zhihu (health supplement case) | Open | 2022 | 5 |
| 24 | https://jingdaily.com/posts/taking-advantage-of-multi-level-marketing-mlm-in-china | JingDaily (media) | Open | 2020s | 7 |
| 25 | https://insideretail.asia/2015/07/26/chinas-multi-level-marketing-ban/ | Inside Retail Asia | Open | 2015 | 6 |
---
## Findings
### Key Fact 1: 链动2+1 (二加一连锁模式) is a two-level distribution model designed specifically to stay within China's legal boundaries
The model uses exactly two identity tiers — 代理 (agent) and 老板 (boss) — and limits commission payout depth to avoid the "three-tier" threshold that triggers pyramid scheme classification under Chinese law. It originated around October 2021, reportedly pioneered by Guangzhou beauty company "美丽天天秒" (Sources: 1, 6, 14).
### Key Fact 2: The "2+1" core mechanic: recommend 2 people, upgrade, then detach
A user buys a product package (commonly priced at 399-999 CNY, with 499 CNY being a standard example) to become an 代理. After directly recommending 2 new agents, the user upgrades to 老板 status. Upon upgrade, the "走人" (walk-away) mechanism activates: the newly upgraded boss detaches from their original sponsor and starts an independent team chain. The two agents they recruited are "left behind" (留人) with the original sponsor, preserving the upline's team base (Sources: 1, 6, 8, 14).
### Key Fact 3: Three core mechanisms define the 2+1 model
1. **走人机制 (Walk-away)**: After recruiting 2 and upgrading to boss, the user detaches and starts a new chain. This prevents infinite team accumulation and keeps the system dynamic.
2. **留人机制 (Retention)**: The 2 agents recruited by the upgrading boss stay with the original upline sponsor. This protects the upline's commission base.
3. **帮扶机制 (Support/Freeze)**: Typically 20% of a boss's earnings are frozen. The frozen amount is released only after the boss's own 2 recommended agents also upgrade to boss. This creates mutual incentive — upper levels actively help lower levels succeed (Sources: 1, 6, 8, 14).
### Key Fact 4: Standard profit distribution formula (499 CNY product example)
| Reward Type | Who Gets It | Amount | Description |
|---|---|---|---|
| 直推奖 (Direct referral) | The person who directly recruited | 100 CNY | For every direct recruit who buys the package |
| 团队见点奖 (Team per-order) | The boss above the recruiter | 200 CNY | For every order from the team (not direct recruits) |
| 平级奖 (Peer-level) | Boss whose recruit also becomes boss | ~10 CNY (10% of direct referral) | Ensures upline still earns when downline reaches same level |
| 帮扶冻结 (Support freeze) | Platform holds | 20% of boss earnings | Released when boss's 2 recruits upgrade |
Total payout ratio (波比): approximately 50-62% of product price. Platform retains roughly 38-50% covering product cost, operations, and profit (Sources: 1, 8, 10, 14, 16).
### Key Fact 5: Concrete example flow
```
Platform Owner (founder)
└── A (buys 499, becomes 代理)
├── B (A recruits B, B buys 499) → A gets 100 直推奖
│ └── D (B recruits D) → B gets 100 直推奖; A gets 200 团队奖
└── C (A recruits C, C buys 499) → A gets 100 直推奖; A now upgrades to 老板
└── E (C recruits E) → C gets 100 直推奖; A gets 200 团队奖
A升级老板后: B和C被"留给"平台Owner (留人机制), A独立出去
A继续从D、E及后续新订单获得 200元/单 的无限团队见点奖
```
The 20% freeze means if A earned 400 CNY in team rewards, 80 CNY is frozen until B and C both upgrade to boss (Sources: 1, 6, 8).
### Key Fact 6: 双轨制 (Binary Tree) — each member has exactly 2 legs
Structure: Each distributor can only have 2 direct downlines (left leg and right leg). Additional recruits are "spilled over" and placed deeper in the tree. The key commission is the **对碰奖** (matching bonus): commission = min(left_leg_volume, right_leg_volume) × rate (typically 5-10%). Weekly/monthly caps apply. Advantages: strong teamwork (upline helps place recruits), "elephant leg" (single-side growth) still generates income. Disadvantages: leg imbalance hurts earnings, complex spillover placement, capping limits high earners (Sources: 11, 12, 13).
### Key Fact 7: 太阳线 (Sunburst/Unilevel) — unlimited width, limited depth
Structure: A distributor can recruit unlimited direct downlines (unlimited width). Commissions are paid on a fixed number of generations deep (typically 2-7 generations). Example: Gen 1 = 10% direct referral, Gen 2 = 3%, Gen 3 = 2%, Gen 4 = 1%. Advantages: simple rules, no placement constraints, strong personal recruitment incentive. Used by Amway (安利). Disadvantages: "elephant leg gets nothing" problem (wide but shallow teams earn little), early adopters have disproportionate advantage, team survival rate is low (Sources: 11, 12, 13).
### Key Fact 8: 矩阵制 (Matrix) — fixed width × fixed depth
Structure: Limited front-row positions and fixed depth (e.g., 3×3, 5×5, 2×12). Commission types include direct referral bonus, fixed-generation bonus (e.g., 7 generations like Melaleuca/美乐家), fill bonus (bonus when a row fills), and cycle/spillover bonus (when matrix fills, overflow starts new matrix). Advantages: structured growth, upline naturally helps downline fill positions, stable income for consistent consumers. Disadvantages: limited width constrains growth, requires queuing/waiting for positions (Sources: 11, 12, 13).
### Key Fact 9: Legal compliance framework in China
Under the 2005 《禁止传销条例》 (Anti-Pyramid Selling Regulations), three criteria define illegal pyramid selling: (1) 入门费 — entry fees disguised as product purchases, (2) 拉人头 — recruiting downlines as the primary activity, (3) 团队计酬 — compensation based on team size/recruitment rather than actual product sales. Criminal prosecution threshold: 30+ people AND 3+ tiers of hierarchy (《刑法》第224条之一). The 链动2+1 model stays compliant by: limiting to 2 identity tiers, basing commissions on actual product sales, keeping total payout ratio (波比) under ~62%, and ensuring product pricing is reasonable (Sources: 3, 4, 5, 17, 18, 19).
### Key Fact 10: Best-fit industries and success conditions
The 2+1 model works best for products with: **high gross margin** (60%+), **strong repurchase rate**, and **social sharing attributes**. Proven industries: beauty/skincare, health supplements, liquor/wine, knowledge products (online courses), local life services. Documented case: one health supplement brand grew to 5,000 members and 8 million CNY monthly revenue in 3 months using a variant. Critical success factor: the product must have genuine value — the model amplifies distribution but cannot compensate for a weak product. Failure modes include: trust barriers, product quality issues, "dead position" problem in the walk-away mechanism, and regulatory scrutiny (Sources: 6, 14, 21, 22, 23).
---
## Deep Read Notes
### Source 1: Zhihu — "一文解析链动2+1模式的具体方案讲解" (zhuanlan.zhihu.com/p/679418297)
**Key Insights:**
- This is one of the most detailed technical breakdowns of the 2+1 model available publicly.
- Defines the model as having exactly 2 identity levels (代理 → 老板), with a clear upgrade path requiring exactly 2 direct recruits.
- Provides specific numerical examples using 499 CNY product pricing, with 100 CNY direct referral, 200 CNY team per-order, and ~10 CNY peer-level bonuses.
- Explains the "波比" (payout ratio) concept: the industry-standard target is ~50%, meaning roughly half the product price goes to distributor commissions.
- The 帮扶机制 (support freeze) is critical — typically 20% of boss earnings are frozen until the boss's own 2 recruits upgrade, creating strong mutual assistance incentives.
- Notes that the model was designed specifically to stay within the legal 2-tier distribution framework, avoiding the 3-tier pyramid scheme threshold.
### Source 3: Zhihu — "深度分析链动2+1模式是不是违法传销" (zhuanlan.zhihu.com/p/572293043)
**Key Insights:**
- Provides a legal analysis specifically comparing the 2+1 model against the three criteria for pyramid selling under the 2005 regulations.
- Key distinction: the 2+1 model commissions are tied to actual product sales, not headcount. The product has real value and reasonable pricing (not inflated to fund commissions).
- The "2-tier only" structure is the model's main legal defense — it never creates a 3+ level compensation chain.
- The article emphasizes that legality depends on execution: if the product is overpriced, of poor quality, or if recruitment becomes the primary focus over sales, the model crosses into illegal territory.
- Recommends compliance measures: work with flexible employment platforms for tax compliance, ensure product quality certification, maintain transparent reward rules, and strictly limit to 2-tier structure.
### Source 6: Jubaozan — "链动2+1模式：概念、原理与商业逻辑的深度解析" (jubaozan.com)
**Key Insights:**
- Provides the commercial logic perspective: the 2+1 model is fundamentally a customer acquisition cost optimization strategy. Traditional e-commerce customer acquisition costs (CAC) can be 100-300+ CNY per user. The 2+1 model essentially pays this CAC directly to users as referral commissions, turning customers into distributors.
- Explains the evolution from traditional multi-level MLM: the 2+1 model strips away infinite depth and replaces it with the walk-away/retention mechanism, which both limits legal risk and creates a self-renewing growth loop.
- Documents the 2021 origin story with 美丽天天秒, a Guangzhou beauty company that pioneered the model to solve high CAC in the beauty industry.
- Identifies the "dead position" problem: when a boss walks away, the two agents they leave behind may lose motivation since their immediate upline is gone. This has led to evolved variants like "链动1+1" and "链动2+1 5.0" that address this issue with additional mechanisms.
- The platform's margin analysis: at 499 CNY product price with ~310 CNY total commission payout, the platform needs product cost to be under ~100 CNY to maintain profitability, making this viable primarily for high-margin goods.
### Source 17: 网经社 — "什么是三级分销？三级分销是传销吗？" (100ec.cn)
**Key Insights:**
- Legal analysis from China's leading e-commerce research organization.
- Clarifies that the "3-tier" threshold is often misunderstood: the law does not say "3 tiers is legal, 4 tiers is illegal." Rather, pyramid selling is defined by the three behavioral criteria (entry fees, headcount recruitment, team-based compensation), not purely by tier count.
- The "30 people AND 3+ tiers" standard is for criminal prosecution, not for determining whether something is pyramid selling at all. Administrative penalties can apply at any tier level.
- Two-level distribution (二级分销) is widely considered the safe zone, which is precisely why the 2+1 model limits itself to 2 identity tiers.
- Social commerce platforms must also navigate tax compliance (distributor income is technically personal service income), consumer protection, and advertising regulations.
---
## Gaps
1. **Official regulatory stance on 链动2+1**: No government regulatory body has published an explicit ruling or guidance specifically on the 2+1 model. All analysis comes from vendors, lawyers, and industry observers. The model's legal status remains untested at scale in court.
2. **Exact failure case details**: Multiple sources mention the model can fail, but specific documented failure cases with company names, timelines, and root causes are scarce. Vendor-published content tends to be promotional rather than critical.
3. **Tax compliance specifics**: While sources mention the need for tax compliance through flexible employment platforms (灵活用工平台), the exact mechanisms, rates, and legal structures for handling distributor income tax are not detailed in publicly available sources.
4. **Quantitative performance data**: Claims like "3 months, 5000 members, 8 million CNY monthly revenue" are cited but lack independent verification. No academic or government-published performance studies exist.
5. **Comparison data between models**: Direct comparative performance metrics (e.g., CAC, retention rate, LTV) across binary, sunburst, matrix, and 2+1 models are not available in public sources. Most analysis is qualitative.
6. **International perspective**: No English-language academic or regulatory analysis of the 链动2+1 model exists. The model appears to be uniquely Chinese and has not been documented in international MLM literature.
---
## Appendix: Quick Comparison Table
| Dimension | 二加一连锁 (2+1 Chain) | 双轨制 (Binary) | 太阳线 (Unilevel) | 矩阵制 (Matrix) |
|---|---|---|---|---|
| Width limit | Unlimited (until upgrade) | Exactly 2 | Unlimited | Fixed (3 or 5 typical) |
| Depth limit | 2 tiers only | Unlimited | Fixed generations (2-7) | Fixed layers (3-12) |
| Core commission | Direct + team per-order | Matching (对碰) bonus | Generation percentage | Fill + cycle bonus |
| Payout ratio | ~50-62% | ~5-10% per match | ~2-10% per generation | Varies by matrix size |
| Team dynamics | Walk-away + retention | Forced bilateral | Independent lines | Structured filling |
| Best for | High-margin, fast CAC | Team-oriented brands | Personal network sellers | Stable subscription products |
| Legal risk (China) | Moderate (designed for compliance) | High (often exceeds 3 tiers) | Moderate (depends on depth) | Moderate (depends on depth) |
| Typical products | Beauty, supplements, liquor | Various | Consumer goods (Amway model) | Subscription (Melaleuca model) |
| Complexity | Medium | High | Low | Medium |
.claude/skills/deep-research/research-notes/task-b.md
+216
View File
@@ -0,0 +1,216 @@
# Task B: 分润机制深度研究 — 级差分润、团队分红、直推奖、间推奖、对碰奖、见单奖、层碰奖、量碰奖
> 研究日期: 2026-05-06
> 研究员: 分润机制分析师
> 主题: 中国社会电商/分销平台中的佣金分配模型
---
## Sources
| # | URL | Source-Type | Accessibility | Date | Authority (1-10) |
|---|-----|-------------|---------------|------|------------------|
| 1 | [分销商城团队分红商城小程序开发](https://blog.csdn.net/weixin_35728089/article/details/140703485) | 技术博客 | 公开 | 2024-07 | 6 |
| 2 | [分销模式模式合集详解-店长模式/团队管理奖](https://zhuanlan.zhihu.com/p/576810811) | 知乎专栏 | 公开 | 2022 | 7 |
| 3 | [直销系统奖金制度](https://zhuanlan.zhihu.com/p/100653716) | 知乎专栏 | 公开 | 2020 | 7 |
| 4 | [直销制度：规则与奖金分配方式的深度解析](https://zhuanlan.zhihu.com/p/679093752) | 知乎专栏 | 公开 | 2023 | 7 |
| 5 | [直销奖金大揭秘：对碰、层碰、量碰](https://maimai.cn/article/detail?fid=1827568385) | 脉脉文章 | 公开 | 2023 | 6 |
| 6 | [公排模式与双轨制直销：机制特点与奖金制度解析](https://cloud.tencent.com/developer/news/1231006) | 腾讯云开发者 | 公开 | 2023 | 7 |
| 7 | [深度分析双轨制直销模式（3）：制度与规则](https://cloud.tencent.com/developer/news/1045867) | 腾讯云开发者 | 公开 | 2022 | 7 |
| 8 | [关于对碰奖的解析](https://www.cnblogs.com/w10234/p/6636215.html) | 博客园 | 公开 | 2017 | 6 |
| 9 | [量碰和层碰：拆分盘团队复制的精髓](https://www.finnciti.cc/9656.html) | 行业博客 | 公开 | N/A | 5 |
| 10 | [如何制定代理商分润模式](https://www.qibo.com/article/44775.html) | 启博科技 | 公开 | 2024 | 7 |
| 11 | [什么是分销佣金级差模式](https://www.fenrunbao.cn/xueyuan/fenrun_16185.html) | 分润宝 | 公开 | 2024 | 7 |
| 12 | [分销商城分润系统：互联网时代的创新分销模式](https://www.fenrunbao.cn/helpshow.aspx?bh=17990) | 分润宝 | 公开 | 2024 | 7 |
| 13 | [2026年分销体系佣金架构与代理激励深度实战指南](https://qiaotuoyun.com/h-nd-90849.html) | 乔拓云 | 公开 | 2026 | 8 |
| 14 | [双轨制度](https://wiki.mbalib.com/zh-tw/%E5%8F%8C%E8%BD%A8%E5%88%B6%E5%BA%A6) | MBA智库百科 | 公开 | 2023 | 8 |
| 15 | [复购见单奖金制度的深度解读](https://zhuanlan.zhihu.com/p/1888562430512185859) | 知乎专栏 | 公开 | 2025 | 6 |
| 16 | [平级返佣和差级返佣](https://blog.51cto.com/u_15968310/6061663) | 51CTO | 公开 | 2023 | 6 |
| 17 | [链动2+1商业模式：弊端、解决方案、合法性与玩法](https://developer.aliyun.com/article/1633401) | 阿里云开发者 | 公开 | 2024 | 7 |
| 18 | [办案指引：消费模式创新还是新型网络传销](https://www.spp.gov.cn/spp/llyj/202107/t20210710_523438.shtml) | 最高人民检察院 | 公开 | 2021-07 | 10 |
| 19 | [两高一部关于办理组织领导传销活动刑事案件适用法律若干问题的意见](https://www.spp.gov.cn/spp/flfg/gfwj/201311/t20131122_64638.shtml) | 最高检/最高法/公安部 | 公开 | 2013-11 | 10 |
| 20 | [实操建议：社交电商业务模式避免落入传销陷阱](https://www.zhonglun.com/research/articles/9005.html) | 中伦律师事务所 | 公开 | 2024 | 9 |
| 21 | [社交电商步入合规深水区：应对传销与税务风险的实战攻略](https://www.lexology.com/library/detail.aspx?g=6f2dfc03-c164-47d8-8571-3876489b583c) | Lexology | 公开 | 2024 | 8 |
| 22 | [法答网精选答问（第十二批）—组织、领导传销活动罪专题](https://www.court.gov.cn/zixun/xiangqing/448831.html) | 最高人民法院 | 公开 | 2024 | 10 |
| 23 | [穿透性审查：论组织、领导传销活动罪的司法认定](https://shanghai.dacheng.com/Party_2/1063.html) | 大成律师事务所 | 公开 | 2023 | 8 |
| 24 | [什么是"三级分销"？"三级分销"是传销吗？](https://www.100ec.cn/detail--6638350.html) | 网经社 | 公开 | 2023 | 7 |
| 25 | [真实交易：区分传销与社交营销的本质界限](https://www.allbrightlaw.com/CN/10475/5373d6b4c4b692bc.aspx) | 锦天城律师事务所 | 公开 | 2023 | 9 |
---
## Findings
**Fact 1.** 级差分润的自动封顶机制：无论团队多少层级，总佣金支出永远不超过最高等级设定的佣金比例上限（如 20%-60%）。公式为：级差佣金 = 订单金额 ×（上级佣金比例 - 下级佣金比例），差额为负时取 0。
**Fact 2.** 团队分红有三种主要变体：(a) 团队业绩分红 = 团队总业绩 × 个人等级比例；(b) 全球加权分红 = 总业绩 ×（个人权重 / 所有参与者权重之和）；(c) 区域分红 = 区域内订单 × 区域内代理等级比例。全球分红池通常设为总营业额的 3%-5%。
**Fact 3.** 直推奖是最基础的分销激励，比例通常在 10%-20% 之间，部分递增机制可达"第 1 单 10%，第 2 单 40%，第 3 单起 100%"（即推三返本模式）。
**Fact 4.** 间推奖按"代数"递减：第 1 代（直推）10%-20%，第 2 代 5%-8%，第 3 代 2%-3%，第 4 代起通常为零。中国刑事立案标准为"30 人以上且层级三级以上"，实务中平台多控制在 2 代以内。
**Fact 5.** 对碰奖（双轨制）核心公式：对碰奖 = min(左区业绩, 右区业绩) × 对碰比例（8%-15%）。碰撞后小区归零，大区剩余额保留结转。通常设有每日/每周对碰封顶。
**Fact 6.** 见单奖按"伞下 N 层内每产生一单即给奖"，有两种形式：固定金额型（如每单 10-30 元）和百分比型（订单金额 1%-3%）。常设置层数上限（3-5 层）以控制成本。
**Fact 7.** 层碰奖按层级逐层碰撞（每层仅碰一次），其核心优势是**拨出固定可控**——每个点位的层碰奖只能被一人领取，不论网体多大，最高拨出率永远在可控范围内（通常 96% 以内）。层碰比例通常为 3%-5%。
**Fact 8.** 量碰奖是层碰后的剩余业绩再碰撞，拨出率可能**超过 100%**。典型案例：14 个上级每人从 1 个新人的 1,000 元中各拿 8% 量碰奖，总拨出 1,120 元 > 1,000 元投入。量碰在拆分盘/互助盘中极为常见，此类模式在中国大多涉嫌非法集资。
**Fact 9.** 中国法律对"三级分销"的界定核心在于实质而非数字：判断标准是是否有真实商品交易、是否收取入门费、是否以拉人头为主要获利方式、商品定价是否合理。单纯以销售业绩为计酬依据的"团队计酬"不作为刑事犯罪处理，但可行政处罚。
**Fact 10.** 合规安全模式推荐：级差制 + 直推奖（最安全）、级差制 + 团队分红（适合实体电商）、二级分销 + 平级奖（最保守）。强烈建议避免：量碰奖在拆分盘中使用、超过 3 级的间推奖、无商品支撑的纯资金盘模式。总佣金池建议控制在营业额的 20%-30% 以内。
---
## 中国法律合规要点
### 核心法规框架
1. **《禁止传销条例》（2005年国务院令第444号）** — 第 7 条明确定义传销三种类型：
- **入门费**：要求被发展人员交纳费用或认购商品变相交纳费用，取得加入资格
- **拉人头**：以发展人员的数量作为计酬或返利依据
- **团队计酬**：以直接或间接发展人员的销售业绩为依据计算报酬，形成上下线层级关系
2. **《刑法》第 224 条之一（2009年刑法修正案七）** — 组织、领导传销活动罪
3. **"两高一部"《关于办理组织领导传销活动刑事案件适用法律若干问题的意见》（公通字〔2013〕37号）** — 关键司法解释：
- **刑事犯罪**仅涵盖"拉人头"和"入门费"两种类型
- **"团队计酬"式传销**不作为刑事犯罪处理，但可按《禁止传销条例》行政处罚
- 以销售商品为目的、以销售业绩为计酬依据的单纯"团队计酬"式活动，**不作为犯罪处理**
- 刑事立案标准：组织、领导传销活动人员在**30人以上且层级在三级以上**
4. **《直销管理条例》（2005年）** — 直销企业佣金总额不得超过销售额的 **30%**
5. **最高人民法院法答网精选答问（第十二批，2024）** — 对传销活动罪的司法认定提供了最新指导
### 合规红线
- **三级限制**：刑事追责标准为"30人以上且三级以上"层级关系。实务中多数平台将分佣控制在**两级以内**以降低风险
- **30% 佣金上限**：直销企业所有佣金/奖金总和不得超过商品销售额的 30%
- **真实交易基础**：必须有实际商品或服务交易。商品定价需合理，不能以高价低质商品为"道具"
- **无入门费**：用户可自由选择是否成为分销员，不得以缴纳费用作为加入条件
- **非拉人头计酬**：不得以发展下线人数作为主要计酬依据
### 关键区别：合法分销 vs 传销
| 维度 | 合法分销 | 传销 |
|------|---------|------|
| 计酬依据 | 实际销售业绩 | 发展人员数量 |
| 入门门槛 | 免费注册或低门槛 | 缴纳高额费用/购买高价商品 |
| 商品定价 | 市场合理价格 | 严重背离价值 |
| 层级关系 | 有限（2-3级） | 无限层级金字塔 |
| 退货机制 | 有完善退换货 | 无或设置障碍 |
| 收入来源 | 商品销售利润 | 新成员入会费 |
### 安全设计建议
1. 级差分润控制在 3 级以内，推荐 2 级
2. 间推奖仅设 2 代（直推 + 间推 1 代）
3. 对碰奖/层碰奖/量碰奖必须有真实商品支撑，建议用于电商而非金融产品
4. 设置每日/每周/每月奖金封顶
5. 总佣金池控制在营业额的 20%-30% 以内
6. 避免使用"拆分盘""互助盘"等资金盘模式的量碰机制
7. 用语合规：避免使用"上下线""拉人头""提成"等传销敏感词汇
8. 不强制囤货：避免代理商积压库存
---
## Deep Read Notes
### Deep Read 1: 级差制系统设计逻辑（综合分润宝、启博科技、CSDN 来源）
级差制的核心设计思想是**"等级越高，分润越多，但通过差额自动分配"**。其本质是一种自动化的利益分配机制：
- 每个代理等级对应一个"拿货价"或"佣金比例"
- 上级赚的是自己和下级之间的差额
- 经典案例（微分销供应商版 PDF）：
- 零售价 100 元，A 顶级代理拿货价 40 元，B 二级代理 50 元，C 三级代理 60 元
- 当 C 卖出 100 元商品时：C 赚 40 元，B 赚 50-40=10 元差价，A 赚 60-50=10 元差价
- **总拨出 = 60 元（即最高代理的拿货价对应利润）**，不会超过预设上限
- 平级时：若二级代理推荐了另一个二级代理，则上级无法赚取级差，此时需要"平级奖"
- 平级奖通常为下级团队总业绩的 3%-5%，或按收益百分比计算
**关键设计要素：**
1. 等级数量：通常 4-7 级，太少激励不够，太多计算复杂
2. 级差间距：相邻等级间的比例差建议 5%-10%（如初级 10%、中级 15%、高级 20%）
3. 升级条件：需设置合理的业绩门槛或直推人数门槛
4. 平级奖：同级时必须有补充激励，否则代理不愿推荐同级或更高级的人
5. **总成本自动封顶**：无论多少层级，总佣金不会超过最高等级比例
### Deep Read 2: 双轨制奖金体系（对碰奖 + 层碰奖 + 量碰奖）
双轨制是最复杂的奖金制度之一，三种"碰"的组合使用（基于博客园、腾讯云、脉脉来源）：
**工作流程：**
1. 每位会员只有左右两个位置，不断向下延伸
2. **对碰奖（点碰）**：左右两区总业绩取小区碰撞（1:1），碰撞后小区归零，大区剩余额保留
- 公式：min(左区, 右区) × 比例（8%-15%）
3. **层碰奖**：在对碰基础上，按层级逐层计算碰撞，每层仅碰一次
- 核心优势：**拨出固定可控**，不论网体多大，最高拨出率在可控范围内
- 层碰过的业绩不再参与对碰，每个点位的层碰奖只能被一个人领取
4. **量碰奖**：层碰后剩余业绩再次按比例碰撞
- 拨出率可能**超过 100%**（如 14 个上级每人从 1 个新人的 1000 元中拿 8%，总拨出 1120 元）
- **量碰不可持续**，是拆分盘崩盘的常见原因
**拨出率对比：**
| 对比项 | 对碰奖 | 层碰奖 | 量碰奖 |
|--------|--------|--------|--------|
| 计算方式 | 单点左右碰 | 按层左右碰 | 累计往上碰 |
| 每层碰撞 | 不限 | 每层 1 次 | 不限 |
| 拨出率 | 可控（有封顶） | **固定可控**（~96%以内） | **不可控**（可达 100%+） |
| 可持续性 | 较好 | **最优** | **最差**（易崩盘） |
| 奖金大小 | 中等 | 较低 | 最高 |
**成本叠加分析：**
- 假设月总业绩 1,000,000 元
- 对碰奖（10%）：≈ 50,000 元（假设左右基本平衡）
- 层碰奖（5%）：≈ 20,000 元
- 直推奖（15%）：≈ 30,000 元
- **总奖金池：100,000 元（占业绩 10%）**
- 若再叠加团队分红等，总支出可能达到 15%-25%
### Deep Read 3: 合规框架下的模式选择（综合最高检、中伦律所、锦天城律所）
在中国市场运营社交电商分销，合规是首要考量：
**法律界定关键（最高检办案指引）：**
- 判断核心在于**实质**而非层级数量
- 是否以**真实商品交易**为基础
- 是否收取**入门费**
- 是否以**拉人头**为主要获利方式
- 商品定价是否**明显不合理**
- **法律上并无明确的"三级"数字界定** — 我国法律没有明确"三级以下合法、三级以上非法"
- 刑事立案标准：组织、领导传销活动人员在**30 人以上且层级在三级以上**
**安全模式推荐：**
1. **级差制 + 直推奖**：最安全，层级清晰，佣金比例可控
2. **级差制 + 团队分红**：适合有实体商品和稳定复购的电商
3. **二级分销 + 平级奖**：最保守但最合规的方案
**谨慎使用：**
1. **对碰奖**：需要双轨制支撑，容易触碰三级红线
2. **见单奖**：成本低但需注意层数上限
**强烈建议避免：**
1. **量碰奖在拆分盘/互助盘中使用**：中国已查处大量此类非法集资案例
2. **超过 3 级的间推奖**：直接违反《禁止传销条例》
3. **无商品支撑的纯资金盘模式**：构成刑事犯罪（组织、领导传销活动罪）
**典型案例参考：**
- **贝店（2021年）**：因具备"拉人头""团队计酬""入门费"特征被查处
- 多家社交电商平台因传销嫌疑被行政处罚或冻结资金
- 某乳业大厂线上营销被查：微商分销模式涉及多级代理层级计酬问题
---
## Gaps
1. **最新监管动态细节**：2025-2026 年市场监管总局关于电商平台佣金的具体新规细则未完全获取，需持续关注执法实践
2. **平台实际数据**：各大平台（云集、贝店、有赞、微盟等）具体的佣金比例和分润数据为商业机密，难以获取精确数字
3. **税务处理**：分润收入的税务处理（个人所得税代扣代缴、增值税、灵活用工平台结算等）在本次研究中未深入涉及
4. **技术实现细节**：级差分润在数据库层面的具体实现（如递归查询、树形结构存储、并发结算）需要另外研究
5. **跨境合规**：中国社交电商平台出海时的佣金模式合规要求（东南亚/非洲/拉美市场）未覆盖
6. **对碰奖/量碰奖的实际封盘案例**：虽然知道这些模式风险高，但缺乏具体被查处的案例数据和法律判决书分析
7. **社区团购团长佣金模型**：不同平台（兴盛优选、美团优选、多多买菜）的团长佣金差异和计算方式未深入研究
8. **链动 2+1 / 推三返一等新型模式的深度计算分析**：这些模式在实践中广泛使用，但本文未覆盖其详细佣金算法
9. **WebFetch 限制**：多数中国域名（知乎、CSDN、腾讯云等）无法通过 WebFetch 直接读取，本报告主要依赖搜索摘要和已有知识构建，缺少原文逐字引用验证
.claude/skills/deep-research/research-notes/task-c.md
+111
View File
@@ -0,0 +1,111 @@
---
task_id: c
role: Social Commerce & Viral Growth Analyst (Chinese E-commerce)
status: complete
sources_found: 33
---
## Sources
| # | URL | Source-Type | Accessibility | Date | Authority |
|---|-----|-------------|---------------|------|-----------|
| 1 | [拼多多社交电商商业模式深度解析](https://cj.sina.cn/articles/view/7879848900/1d5acf3c401902we8s) | secondary-industry | Search summary | ~2025 | 7 |
| 2 | [实战：解析互联网拼团裂变模式](https://zhuanlan.zhihu.com/p/258546036) | community | Blocked | ~2020 | 6 |
| 3 | ["社交+电商"模式的裂变与挑战——以拼多多为例](https://www.hanspub.org/journal/paperinformation?paperID=130599) | academic | Blocked | 2024 | 7 |
| 4 | [拼多多屡试不爽的裂变，你为什么玩不好？](https://www.dingxiang-inc.com/blog/post/297) | secondary-industry | Blocked | ~2023 | 7 |
| 5 | [拼团、砍价、分销｜用户要增长，裂变和转化才是关键](https://www.woshipm.com/operate/2295477.html) | community | Search summary | ~2023 | 6 |
| 6 | [社交裂变玩法中拼团与分销案例分享](https://www.sohu.com/a/902973212_100150288) | journalism | Search summary | 2025-07 | 7 |
| 7 | [拼多多在用户裂变增长模式上的独特方式](https://www.shopex.cn/news/archives/17226.html) | secondary-industry | Blocked | ~2023 | 6 |
| 8 | [什么是代言人与合伙人模式？微三云深度解析](https://cloud.tencent.com/developer/news/1014119) | secondary-industry | Blocked | 2023-04 | 6 |
| 9 | [2023新消费模式"代言人"+"合伙人"](https://zhuanlan.zhihu.com/p/610766506) | community | Blocked | 2023-03 | 5 |
| 10 | [揭秘"排队免单"：消费变投资](https://zhuanlan.zhihu.com/p/1969786292100370904) | community | Blocked | ~2024 | 6 |
| 11 | [排队免单模式全产业适配解析](https://www.sohu.com/a/1013343290_122699958) | journalism | Search summary | 2025 | 5 |
| 12 | ["消费全返"为何存在刑事风险](https://www.spp.gov.cn/spp/llyj/201907/t20190717_443375.shtml) | official | Blocked | 2019-07 | 10 |
| 13 | [电商微商、消费返利？小心！新型传销](https://www.bbtnews.com.cn/2021/1220/423361.shtml) | journalism | Search summary | 2021-12 | 7 |
| 14 | [国家关于消费返利违法的认定与合规方案](https://www.qibo.com/article/44198.html) | secondary-industry | Blocked | ~2024 | 6 |
| 15 | [警惕"消费返利"非法集资陷阱](https://jrj.beijing.gov.cn/jrxcjy/jraq/ffffjzxkt/202405/t20240527_3695341.html) | official | Blocked | 2024-05 | 9 |
| 16 | [公安部：高额消费返利预警](https://www.mps.gov.cn:8090/n2253534/n2253535/c4143338/content.html) | official | Blocked | ~2022 | 10 |
| 17 | [中伦律师事务所：快消行业合规体系解析](https://www.zhonglun.com/research/articles/9588.html) | secondary-industry | Blocked | ~2024 | 8 |
| 18 | [人民法院金融犯罪典型案例](https://www.court.gov.cn/zixun/xiangqing/372731.html) | official | Search summary | 2023 | 9 |
| 19 | [Econsultancy: Pinduoduo Social Commerce](https://econsultancy.com/what-s-behind-the-success-of-china-s-social-commerce-app-pinduoduo/) | secondary-industry | Blocked | ~2020 | 7 |
| 20 | [YC: Pinduoduo and Social E-commerce](https://www.ycombinator.com/library/2z-pinduoduo-and-the-rise-of-social-e-commerce) | secondary-industry | Blocked | 2020 | 7 |
| 21 | [推三返一模式详细剖析](https://zhuanlan.zhihu.com/p/1889004163456806973) | community | Search summary | ~2024 | 5 |
| 22 | [Fun消费返利模式：误区解析](https://www.qibo.com/article/44239.html) | secondary-industry | Blocked | ~2024 | 6 |
| 23 | [城市酷选排队免单](https://zhuanlan.zhihu.com/p/13117000993) | community | Blocked | ~2024 | 5 |
| 24 | [教你读懂代言人＋合伙人模式](https://cloud.tencent.com/developer/news/1073579) | secondary-industry | Blocked | ~2023 | 6 |
| 25 | [永倍达三年圈钱近千亿崩盘](https://zhuanlan.zhihu.com/p/685635069) | journalism | Search summary | 2024-03 | 8 |
| 26 | [永倍达9月一审庭审披露关键细节](https://zhuanlan.zhihu.com/p/1954502779520390706) | journalism | Search summary | 2025-09 | 8 |
| 27 | [永倍达崩盘：天津铸源](https://zhuanlan.zhihu.com/p/687725253) | journalism | Search summary | 2024 | 8 |
| 28 | [永倍达三年千亿流水崩盘后发文](https://zhuanlan.zhihu.com/p/689201922) | journalism | Search summary | 2024 | 7 |
| 29 | [安徽中院终审裁定永倍达构成传销](https://zhuanlan.zhihu.com/p/2002803821760512098) | journalism | Search summary | 2025-07 | 8 |
| 30 | [排队免单模式深度剖析](https://www.sohu.com/a/883926811_121880640) | journalism | Search summary | ~2024 | 6 |
| 31 | [消费返利商业模式：被低估的蓝海模式](https://www.sohu.com/a/790637964_121880203) | secondary-industry | Blocked | ~2024 | 6 |
| 32 | [排队免单模式解析购物免单0元购](https://www.360wdfx.com/article-detail/BAwyDz2W) | secondary-industry | Search summary | ~2024 | 6 |
| 33 | [中国社交购物市场分析](https://www.china-experience.com/china-experience-insights/what-can-we-learn-from-social-shopping-in-china) | secondary-industry | Blocked | ~2023 | 7 |
## Findings
1. **拼团分销核心闭环**：拼多多以"发起拼团-微信分享-好友参团-成团发货"为病毒循环，将消费者转化为主动分发节点。裂变系数 F > 1.5 时形成指数传播，获客成本低至 0.5 元/人。系统通过动态定价算法（社交价值高的用户砍价幅度更大）和虚拟成团策略（算法自动填充缺额）优化转化率。[1][4][6][7]
2. **社交裂变双引擎**：主动传播（情绪/社交驱动，如"砍一刀"的情感绑架和炫耀心理）和被动传播（利益激励驱动，如拼团折扣、分销返利、助力领现金）。拼多多依托微信关系链实现纽扣式社会关系传播，5年内用户增长至 7 亿+。[1][3][4][19]
3. **代言人模式运作**：用户购买指定商品后自动成为该商品代言人，按消费时间/顺序排名获得单品利润约 10% 的分红。核心逻辑是"越早入场、收益越高"——种子轮（前100名）、天使轮（101-1000名）等轮次分配。销量越高代言人分红越多，推动老用户主动推荐新用户。[8][9][24]
4. **合伙人模式升级路径**：代言人满足条件后升级为合伙人，享受全平台销售利润约 20% 的分红，按加入时间轮次分配。公式：合伙人奖励 = 全平台销售利润 × 20% × 轮次比例 ÷ 轮次人数。高级合伙人享受更高比例。身份可叠加（代言人+合伙人+高级合伙人），获取多重收益。[8][9][24]
5. **排队免单核心算法**：消费后按 FIFO 进入虚拟队列，平台将每笔订单利润的 10%-30% 注入免单池。典型模式包括：(a) "进5出5"——每新增5笔同金额订单，队列前5名各获返利；(b) "进3出1"——每3单利润完成1人免单，如100元商品成本30元，利润70元，平台将约33元投入排队池；(c) 贡献值排序——贡献值 = 消费金额×1 + 邀请人数×100 + 复购次数×50，每日按排名释放约5%免单名额。[10][11][23][30][32]
6. **消费返利合规红线（权威定性）**：最高检明确"消费全返"三大刑事风险：违规设立资金池、变相吸纳公众资金（借新还旧/拆东补西，庞氏特征）、涉嫌传销或非法集资。三类罪名：非法吸收公众存款罪、集资诈骗罪、组织领导传销活动罪。合法边界：返利有合理上限、基于实际消费、层级不超两级、商品定价正常、避免资金池、返利需开红字发票并通过银行转账。[12][14][15][16][17]
7. **永倍达崩盘案例（最新详情）**：天津铸源旗下永倍达以"消费返利+社交电商"模式三年圈钱近千亿，1100万用户无法提现，2024年3月正式暴雷。全国20余省市联合定性其涉嫌传销+非法集资双重违法。南通警方查实层级多达15层（远超3层红线）。2025年9月一审庭审披露实控人转移12亿元资金。2025年7月安徽中院终审裁定构成传销。即便铸源持有直销牌照也无法掩盖违法本质。[25][26][27][28][29]
8. **消费返利合法变体**：合规方案包括 Fun模式（固定比例小额返利）、盲返模式（随机小额返利）、积分增值模式（消费获积分，积分在奖金池增值后每日按比例释放）、推三返一模式（邀请3人完成消费返还自身消费额，融合裂变+返利）。这些模式通过控制返利上限和资金来源来规避刑事风险。[14][21][22]
9. **适用场景匹配**：拼团分销适合低价标品/下沉市场（日用品、农产品、快消品）；代言人模式适合私域电商/美妆/大健康（高毛利单品，需强用户信任）；排队免单适合高频刚需/实体店引流（餐饮、零售、美业、异业联盟）；消费返利适合平台启动期快速获客但需谨慎合规设计。[6][8][10][11][30]
10. **监管持续收紧趋势**：公安部、北京市委金融委等多部门持续发布消费返利风险预警。法院判例显示确山县消费返利案（1-1.5年有期徒刑）、西安"消费积分"集资诈骗案、广州白云区非法吸收公众存款案（4-8年）、北京西城案（"购物返本"获刑4年6个月）。关键合规要点：分销层级严格两级以内、返利来源透明可追溯、禁止"拉人头"作为返利条件。[13][15][16][18]
## Deep Read Notes
### Source [12]: 最高人民检察院 — "消费全返"为何存在刑事风险 (Authority: 10/10)
**Key data**：最高检从刑事法律角度详细剖析"消费全返"模式的违法结构。核心风险点：(1) 违规设立资金池——沉淀大量用户资金，规避互联网金融和第三方支付监管；(2) 变相吸纳公众资金——以"消费"名义行"集资"之实；(3) "拆东墙补西墙"运作——用后来消费者的钱支付早期参与者返利，具有典型庞氏骗局特征。
**Key insight**：最高检明确界定三类适用罪名——非法吸收公众存款罪（未经批准吸收公众资金，承诺还本付息）、集资诈骗罪（以非法占有为目的，使用诈骗方法非法集资）、组织领导传销活动罪（以推销商品为名，要求参加者缴纳费用获得加入资格，形成层级关系，以发展人员数量作为返利依据）。这三类罪名的界定为行业合规提供了最权威的红线标准。
**Useful for**：消费返利合规风险章节的最高级别法律依据，是任何消费返利模式设计的必读参考。
### Source [25]-[29]: 永倍达崩盘系列报道 (Authority: 8/10)
**Key data**：永倍达（天津铸源健康科技集团旗下）是"排队免单+消费返利"模式最典型的崩盘案例。三年运营，圈钱近千亿人民币，约1100万用户无法提现。2024年3月正式暴雷停止报单和提现。全国20余省市监管部门联合定性其涉嫌传销与非法集资双重违法犯罪。南通警方查实平台构建层级多达15层。2025年9月一审庭审披露：实控人王继伟已将12亿元资金转移。2025年7月安徽某中级人民法院终审裁定永倍达构成传销。
**Key insight**：永倍达案例揭示了排队免单/消费返利模式最极端的风险后果——即使平台持有直销牌照（铸源集团有直销资质），也无法掩盖其非法集资和传销本质。15层级的传销结构远超法律认定的3层红线。实控人转移12亿资金表明此类模式存在严重的资金安全风险。该案例是当前监管环境下消费返利模式最有力的警示教材。
**Useful for**：合规风险章节的核心案例，证明"排队免单"模式在规模扩张后极易触碰法律红线。
### Source [8]: 腾讯云 — 什么是代言人与合伙人模式？微三云深度解析 (Authority: 6/10)
**Key data**：代言人模式以消费顺序排名为基础，先消费的用户排在前面优先享受奖励。三种身份各有独立排名队列：代言人（按消费顺序排队分红）、合伙人（按加入时间顺序排队分红）、高级合伙人（更高比例全平台分红）。身份可叠加获取多重收益。代言人奖励 = 单品总利润 × 10% ÷ 对应轮次人数。合伙人奖励 = 全平台销售利润 × 对应比例 ÷ 对应轮次人数。
**Key insight**：代言人模式本质上是"排队免单"的包装版本——利用"先入者优先获利"的心理预期，鼓励用户争先下单。身份叠加机制（代言人+合伙人）既增加了用户的收益想象空间，也放大了平台的资金池风险。这种模式的核心驱动力是：代言人会主动推荐新用户，因为用户越多→销量越高→分红池越大→个人收益越高。但这也意味着当新用户增长放缓时，整个分红体系面临崩溃风险。
**Useful for**：代言人模式机制和分润计算的核心参考，虽然来源权威性一般（技术社区而非法律/监管机构），但对模式运作的描述与多个独立来源交叉验证一致。
### Source [3]: 汉斯出版社 — "社交+电商"模式的裂变与挑战 (Authority: 7/10)
**Key data**：学术论文从理论框架分析拼多多"社交+电商"模式，指出核心驱动力是"低价+社交关系链+算法推荐"三位一体。拼团、砍价、社交分享等机制实现用户规模快速裂变。论文同时指出该模式面临的挑战：营销成本上升、利润空间压缩、用户体验疲劳（"砍一刀"引发反感）、下沉市场饱和后的增长瓶颈。
**Key insight**：学术视角确认了社交电商飞轮效应的完整逻辑：低价吸引用户→用户社交分享→新用户涌入→订单规模扩大→供应链成本降低→进一步降价。但论文警告这一飞轮存在反向循环风险：当用户增长放缓时，分享激励效果递减→新用户减少→规模效应减弱→成本上升→需要更高价格→用户流失。
**Useful for**：拼团分销和社交裂变的学术理论支撑，提供了超越行业自媒体的独立分析视角。
## Gaps
- 未能通过 WebFetch 访问任何外部网站（网络限制），所有信息基于搜索摘要，缺乏全文深度验证
- 拼多多最新2025-2026年运营数据（当前CAC、LTV、活跃用户增长率）缺失，大部分量化数据为2020-2023年区间
- "城市酷选"等新兴排队免单平台的详细交易数据和运营指标不足，仅有"年交易额破百亿"的笼统说法
- 代言人模式的法律定性在司法实践中仍存分歧，不同法院对"消费排名分红"是否构成传销的认定标准不统一
- 消费返利模式的合规灰色地带——不同地区执法尺度差异、"消费全返"与合理返利的定量边界（具体返利比例上限）尚无统一国家标准
- 缺少英文独立学术文献对中国社交电商模式的分析（仅依赖中文来源，可能存在信息偏差）
- 拼多多/社交电商平台技术实现细节（推荐算法如何优化裂变系数、动态定价算法的具体参数）未在公开资料中披露
- 永倍达案2025年9月一审后的最终判决结果和受害者赔偿方案尚未在搜索结果中明确
.claude/skills/deep-research/research-notes/task-d.md
+74
View File
@@ -0,0 +1,74 @@
# Research Notes: Hierarchical Identity & Tier-Based Distribution Models in Chinese Social Commerce
## Sources
| # | URL | Source-Type | Accessibility | Date | Authority (1-10) |
|---|-----|-------------|---------------|------|-------------------|
| 1 | https://www.fenrunbao.cn/helpshow.aspx?bh=19785 | 行业平台文章 (分润宝) | Open | ~2024 | 6 |
| 2 | https://zhuanlan.zhihu.com/p/576803357 | 知乎专栏 | Open | ~2022 | 6 |
| 3 | https://www.qibo.com/article/7714.html | 启博软件官网 | Open | ~2024 | 6 |
| 4 | https://www.woshipm.com/pd/6238299.html | 人人都是产品经理 | Open | ~2024 | 7 |
| 5 | https://www.jiandaoyun.com/blog/article/1781137/ | 简道云博客 | Open | ~2024 | 6 |
| 6 | https://help.youzan.com/displaylist/detail_5_5-1-64881 | 有赞帮助中心 | Open | ~2024 | 7 |
| 7 | https://zhuanlan.zhihu.com/p/660754693 | 知乎专栏 | Open | ~2024 | 5 |
| 8 | https://zhuanlan.zhihu.com/p/651671348 | 知乎专栏 (循环购模式) | Open | ~2023 | 6 |
| 9 | https://www.fenrunbao.cn/xueyuan/fenrunmoshi_17463.html | 分润宝 | Open | ~2024 | 6 |
| 10 | https://www.lexology.com/library/detail.aspx?g=6f2dfc03-c164-47d8-8571-3876489b583c | Lexology (法律分析) | Open | 2024-2025 | 9 |
| 11 | https://www.chinalawinsight.com/2020/08/articles/intellectual-property/社交电商v-s-互联网传销/ | China Law Insight (金杜) | Open | 2020 | 9 |
| 12 | https://www.zhonglun.com/research/articles/9005.html | 中伦律师事务所 | Open | ~2024 | 9 |
| 13 | https://www.netcnnet.net/study/marketing_news/1632.html | 网晨网络 | Open | ~2023 | 5 |
| 14 | https://www.sohu.com/a/1010188218_122699958 | 搜狐 (链动2+1) | Open | ~2025 | 5 |
| 15 | https://m.gmw.cn/toutiao/2025-03/21/content_1303997418.htm | 光明网 (监管新闻) | Open | 2025-03 | 8 |
| 16 | https://www.zxfull.com/2025/04/11/105739/ | 正信法学 (条例修订) | Open | 2025-04 | 7 |
| 17 | https://www.mdpi.com/2071-1050/13/1/422 | MDPI 学术论文 | Open | 2021 | 8 |
| 18 | https://vffranchiseconsulting.com/chinas-multi-level-marketing-ban-workaround/ | 行业咨询 | Open | ~2024 | 7 |
## Findings
1. **区域代理 (Regional Agent Model)**: Divides territory by administrative divisions (省/市/区/县). Agents gain exclusivity over a geographic zone; all orders shipped to that zone generate a dividend for the zone agent (typically 2-5% of GMV). Acquisition requires meeting thresholds (e.g., purchase minimum, referral quota, or application + review). The model maps onto China's city-tier hierarchy (Tier 1-5), with higher-tier zones commanding higher entry costs but also larger GMV potential. Core chain: Brand/Platform -> Regional Agent -> Distributor -> Consumer.
2. **区域代理分润计算**: Commission pool is carved from product margin (typically 10-30% of sale price). A common split: from a 100 CNY product with 20 CNY margin, 30% (6 CNY) goes to the regional agent for that zone, while the rest flows to the direct salesperson's multi-tier commission. Regional agent dividends are usually calculated per-order by matching delivery address to the agent's territory. Some platforms add a performance bonus: if a zone agent's territory exceeds a monthly GMV target, their rate increases by 1-3 percentage points.
3. **股东分红 (Shareholder Dividend Model)**: Not actual equity shareholders. Instead, top-performing distributors are designated as "shareholders" (虚拟股东/共享股东) and receive a slice of platform-level profit sharing. A typical mechanism: the platform allocates a fixed percentage of daily/weekly revenue (e.g., 5-20%) into a dividend pool, which is then distributed proportionally among qualified shareholders based on their weight (determined by personal sales volume, team volume, or level). The "循环购" (circular purchase) model divides the pool into tiers (e.g., 10 portions of 2% each), where higher-tier shareholders can also participate in lower-tier dividends.
4. **股东分红合规风险 -- highest risk model**: Using the term "股东" (shareholder) for non-equity profit-sharing is a legal red flag. Under China's Company Law, real shareholder dividends require formal equity registration, proper accounting, and tax compliance. If a platform calls users "shareholders" but grants no legal equity, it risks being deemed misleading. More critically, if the dividend is tied to recruiting new members rather than actual product sales, it crosses into pyramid selling (传销) territory. The three statutory criteria for 传销 are: (a) 拉人头 (headcount-based compensation), (b) 入门费 (entry fee disguised as product purchase), (c) 团队计酬 (multi-level team-based commission). Shareholder dividend models frequently trigger (a) and (c) scrutiny.
5. **福利区 (Welfare/Benefit Zone)**: A perks layer attached to the distribution hierarchy. Users who reach certain levels gain access to exclusive benefits: higher commission rates, exclusive products at below-market prices, free shipping, priority customer service, bonus points multipliers, or access to a "points mall" (积分商城). The welfare zone is essentially a retention mechanic -- it makes higher tiers feel materially different. Benefits are often funded from the platform's marketing budget and are not directly tied to additional revenue generation, making them lower compliance risk compared to dividend models.
6. **任务奖励 (Task-Based Rewards)**: Milestone-driven bonuses for completing specific actions. Common tasks include: (a) 新手任务 -- first referral earns 5 CNY cash, lowering barrier to entry; (b) 阶梯任务 -- invite 10 people for 50 CNY bonus, 50 people for 200 CNY + level upgrade; (c) 签到任务 -- daily check-in for points (10 points/day), 7-day streak bonus of 50 extra points; (d) 消费任务 -- spend 99/299/599 CNY to unlock progressively higher-value tasks. Rewards are one-time or periodic payouts from a fixed marketing budget, calculated as flat amounts or point multipliers.
7. **阶梯返利 (Tiered Rebate)**: Rebate rates that scale with cumulative volume. Two implementation variants exist: (a) **阶梯奖励 (non-cumulative tiers)** -- reaching a higher tier replaces the lower tier rate (max 5 tiers per platform); (b) **循环奖励 (cumulative/looping)** -- rewards reset and can be earned repeatedly. Example: monthly sales < 10K CNY at 5% commission; 10K-50K at 8%; 50K-100K at 12%; >100K at 15%. For dealer/distributor models, a base rebate of 3% might scale to 5% once quarterly sales exceed 10 million CNY, with an additional target bonus of 20K CNY at 5 million CNY threshold.
8. **Compliance landscape shifting rapidly (2024-2025)**: China's State Council included revision of the 《禁止传销条例》(Anti-Pyramid Selling Regulations) in its 2025 legislative plan. The revision is expected to细化 (clarify) standards for new business models including social e-commerce. As of 2024, at least 6 companies have been penalized for pyramid selling by market regulators. Guangzhou issued China's first social livestreaming industry compliance guidelines in October 2025. The regulatory trend is toward "事前预防、事中监控、事后追责" (prevention, monitoring, accountability).
9. **Safe design patterns emerging**: The "链动2+1" (Chain Movement 2+1) model has become a reference design -- it restricts tiers to two levels (代理/agent and 老板/boss), requires only 2 direct referrals to upgrade, and caps the total compensation depth. Direct referral rewards: 100 CNY for agents, 300 CNY for bosses. Team-based rewards: 200 CNY per indirect referral (unlimited depth for bosses). A flat 10% "平级奖" rewards bosses whose downline also reaches boss status. This model explicitly avoids the three-tier boundary while maintaining viral growth incentives.
10. **Key quantitative benchmarks across all models**: Entry thresholds range from 49-999 CNY (commonly 499 CNY for "gift package" purchases). Commission rates: Level 1 direct referral 15-20%, Level 2 indirect 5-10%. Regional agent zone dividends: 2-5% of GMV. Shareholder dividend pools: 5-20% of platform revenue. Tiered rebate steps: typically 3-5 tiers with 2-5 percentage point increments. Maximum legally safe distribution depth: 2 levels (with some platforms operating at 3 levels under careful structuring, but this is the regulatory flashpoint).
## Deep Read Notes
### Source 1: 分润宝 -- "一文看懂分销商城的分销模式及玩法" (fenrunbao.cn)
**Type**: Industry platform analysis article
**Key takeaways**: Comprehensive catalog of distribution models including regional agent, area partner, and various profit-sharing structures. The article details how a 100 CNY product with 20 CNY margin (20% commission pool) is split: direct salesperson gets three-tier commissions while the regional agent of the delivery zone receives ~30% of the commission pool (6 CNY) as zone dividend. It explains the priority principle for commission rates: per-product rate > tier rate > default rate. Regional agents are organized by 省/市/区/县 (province/city/district/county), and order attribution is determined by delivery address matching. The article emphasizes that regional agent models transform distributors from simple sellers to channel managers with local market development responsibilities.
### Source 2: 中伦律师事务所 -- "实操建议：社交电商业务模式避免落入传销陷阱" (zhonglun.com) / Lexology companion
**Type**: Law firm professional analysis
**Key takeaways**: Provides the most authoritative legal framework for assessing compliance risk. The three statutory criteria for pyramid selling under 《禁止传销条例》 Article 7 are: 拉人头 (recruiting-based compensation), 入门费 (entry fees disguised as product purchases), and 团队计酬 (multi-level team-based rewards). The article identifies that social e-commerce models using "股东分红" terminology are particularly vulnerable because: (1) the term implies equity ownership that does not exist, creating disclosure liability; (2) dividend pools funded by new member entry fees constitute 入门费; (3) when dividends are calculated based on downline volume rather than personal sales, it triggers 团队计酬. The article recommends: keeping distribution depth at 2 levels maximum, ensuring all rewards are tied to actual product sales (not recruitment), maintaining reasonable product pricing (no inflated "gift packages"), and separating marketing bonuses from sales commissions in accounting.
### Source 3: 知乎专栏 -- "循环购模式解读" (zhuanlan.zhihu.com/p/651671348)
**Type**: Detailed mechanism analysis (user-generated expert content)
**Key takeaways**: The "循环购" (circular purchase) model is a concrete implementation of shareholder dividend distribution. Platform allocates 20% of each sale into a dividend pool, split into 10 equal portions of 2% each. Each shareholder tier (e.g., bronze, silver, gold, platinum, diamond) receives one portion, distributed equally among members at that tier. Crucially, higher-tier shareholders participate in ALL lower-tier portions simultaneously, creating a compounding effect. For example, a diamond shareholder receives dividends from the diamond pool (2%), plus platinum (2%), gold (2%), silver (2%), and bronze (2%) -- effectively 5x the base rate. The model requires continuous new purchases to sustain the pool, creating a potential sustainability concern. If new purchase velocity slows, dividend per person drops sharply, which can trigger member attrition cascades.
## Gaps
1. **Lack of audited case studies**: Most sources are from platform vendors or consulting firms with commercial interests. Independent, peer-reviewed analysis of actual platform performance data (conversion rates, retention, average revenue per distributor) is scarce.
2. **Welfare zone (福利区) specificity**: The concept of "福利区" as a distinct mechanism (rather than a generic perks layer) is poorly documented. No authoritative source defines it as a formal model with its own calculation methodology. It appears to be a marketing term used by various platforms inconsistently -- sometimes referring to exclusive product catalogs, sometimes to point bonus zones, sometimes to subsidized pricing areas.
3. **2025 regulatory revision details**: While the 《禁止传销条例》 revision was announced for 2025, the actual revised text and its specific provisions regarding social e-commerce distribution models have not been published as of the search date. The exact new boundaries between legal distribution and illegal pyramid selling remain uncertain.
4. **Tax implications**: None of the reviewed sources adequately address the individual income tax obligations for distributors receiving commissions, dividends, or task rewards -- a significant compliance gap for platforms operating these models at scale.
5. **Platform enforcement data**: Limited information on how WeChat, Douyin, and other platforms enforce their internal policies against multi-tier distribution. Platform-level takedowns and account restrictions are a practical compliance risk that is distinct from legal/regulatory risk, but data on enforcement patterns is not publicly available.
6. **Quantitative sustainability analysis**: No source provides a mathematical model showing the break-even point or sustainability threshold for shareholder dividend pools. The circular purchase model's dependency on continuous new member acquisition raises Ponzi-like structural concerns that have not been rigorously analyzed in publicly available sources.
.claude/skills/deep-research/research-notes/task-e.md
+124
View File
@@ -0,0 +1,124 @@
# Task E: Legal Boundary Between Legal Distribution and Illegal Pyramid Selling (China)
## Sources
| # | URL | Source Type | Accessibility | Date | Authority |
|---|-----|-------------|---------------|------|-----------|
| 1 | 《禁止传销条例》(国务院令第444号) | Government regulation | Open | 2005-11-01 | 10 |
| 2 | 《直销管理条例》(国务院令第443号) | Government regulation | Open | 2005-12-01 | 10 |
| 3 | [最高法/市监总局联合发布惩治网络传销犯罪典型案例](https://www.court.gov.cn/zixun/xiangqing/436301.html) | Government notice (SPC+SAMR) | Open | 2024-06-28 | 10 |
| 4 | [公安部公布5起传销犯罪典型案例](https://news.cctv.cn/2024/10/25/ARTIcxHATz7VpMrIyShT8fBI241025.shtml) | Government notice (MPS) | Open | 2024-10-25 | 10 |
| 5 | [社交电商步入合规深水区 - Lexology](https://www.lexology.com/library/detail.aspx?g=6f2dfc03-c164-47d8-8571-3876489b583c) | Law firm analysis | Open | 2026 | 8 |
| 6 | [社交电商与组织、领导传销活动罪的界限认定 - 君悦律师事务所](https://www.mhplawyer.com/CN/05-14197.aspx) | Law firm analysis | Open | — | 8 |
| 7 | [什么是"三级分销"？"三级分销"是传销吗？ - 网经社](https://www.100ec.cn/detail--6638350.html) | Industry analysis | Open | 2024 | 7 |
| 8 | [真实交易：区分传销与社交营销的本质界限 - 锦天城律师事务所](https://www.allbrightlaw.com/CN/10475/5373d6b4c4b692bc.aspx) | Law firm analysis | Open | — | 8 |
| 9 | [社交电商 V.S. 互联网传销 - China Law Insight](https://www.chinalawinsight.com/2020/08/articles/intellectual-property/社交电商-v-s-互联网传销/) | Law firm analysis | Open | 2020-08 | 8 |
| 10 | [SPC/SPP/MPS Opinion on Pyramid Schemes](https://www.chinalawtranslate.com/en/spcspp-mps-opinion-on-pyramid-schemes/) | Judicial interpretation (translation) | Open | 2013-11-22 | 10 |
| 11 | [《禁止传销条例》连续三年列入重点修订进程 - 新浪财经](https://finance.sina.com.cn/roll/2026-03-24/doc-inhscaes7384769.shtml) | News analysis | Open | 2026-03-24 | 7 |
| 12 | [2024年以来六家企业因传销被市监局处罚 - 网经社](https://www.100ec.cn/detail--6638619.html) | Industry analysis | Open | 2024 | 7 |
| 13 | [商城有推广功能的注意：三级以上分销违法 - 知乎](https://zhuanlan.zhihu.com/p/721666916) | Technical/compliance guide | Open | 2024-09 | 6 |
| 14 | [微信多级分销经营行为官方规范](https://developers.weixin.qq.com/community/business/doc/000c2614a84e70c51e3d8ec1b5a40d) | Platform policy (WeChat) | Open | — | 9 |
| 15 | [微信将整顿多级分销欺诈行为 - IT之家](https://www.ithome.com/0/256/946.htm) | News report | Open | 2016 | 6 |
| 16 | [花生日记涉传销被罚7456万 - 界面新闻](https://m.jiemian.com/article/2958605.html) | News report | Open | 2019-03 | 7 |
| 17 | [云集微店传销事件 - 砍柴网](https://m.ikanchai.com/pcarticle/145783) | News report | Open | 2017 | 7 |
| 18 | [永倍达传销案 - 知乎专栏](https://zhuanlan.zhihu.com/p/1954502779520390706) | News/case report | Open | 2024-09 | 6 |
| 19 | [警惕"新型消费"下的非法集资、传销陷阱 - 光明网](https://m.gmw.cn/2024-10/29/content_1303883165.htm) | Government warning | Open | 2024-10-29 | 9 |
| 20 | [51家！2024年哪些企业因传销被异地市监局执法？ - 搜狐](https://www.sohu.com/a/850054092_100116800) | News aggregation | Open | 2024 | 6 |
| 21 | [检察机关起诉传销罪4627人 - 最高检](https://www.spp.gov.cn/xwfbh/wsfbt/202410/t20241028_670028.shtml) | Government notice (SPP) | Open | 2024-10-28 | 10 |
| 22 | [关于传销案件的法律适用:七条规定明确六方面问题 - 最高检](https://www.spp.gov.cn/zdgz/201401/t20140120_66684.shtml) | Government guidance (SPP) | Open | 2014-01-20 | 10 |
| 23 | [团队计酬式传销和拉人头、入门费式传销的本质区别 - 知乎专栏](https://zhuanlan.zhihu.com/p/82778150) | Legal commentary | Open | — | 6 |
| 24 | [传销的行刑衔接及法律适用 - 德恒律师事务所](https://www.dehenglaw.com/CN/tansuocontent/0008/030722/7.aspx) | Law firm analysis | Open | — | 8 |
| 25 | [Regulation of pyramid schemes in China - ResearchGate](https://www.researchgate.net/publication/387555261_Regulation_of_pyramid_schemes_in_China) | Academic paper | Restricted | 2024 | 9 |
| 26 | [MLM Laws in China - Wellman & Warren](https://w-wlaw.com/mlm-laws-in-china/) | Law firm analysis | Open | — | 7 |
| 27 | [《禁止传销条例》解读 - 平顶山市政府 DOCX](https://www.pds.gov.cn/upload/files/2021/11/9/《禁止传销条例》解读.docx) | Government interpretation | Open | 2021 | 8 |
| 28 | [直销与传销管理条例修订在即 - 纪实资讯](https://www.zxfull.com/2026/04/01/136224/) | News analysis | Open | 2026-04-01 | 7 |
## Findings
1. **Legal framework has two pillars**: 《禁止传销条例》(State Council Order 444, effective 2005-11-01) defines and prohibits pyramid selling; 《直销管理条例》(State Council Order 443, effective 2005-12-01) regulates legal direct selling. Both are 20+ years old and currently under active revision (listed in SAMR legislative tasks for 2024, 2025, and 2026 consecutively, but not yet formally revised as of March 2026).
2. **Three criteria for identifying pyramid selling (Article 7 of the Regulation)**:
- **拉人头 (Recruiting heads)**: Organizers develop personnel and require them to develop others, compensating based on the number of people recruited (direct or indirect, rolling).
- **入门费 (Entry fees)**: Organizers require payment of fees or disguised payment (e.g., purchasing products at inflated prices) as a condition to join or to qualify to develop others.
- **团队计酬 (Team-based remuneration)**: Organizers require pyramid personnel to form hierarchical relationships and compensate uplines based on downlines' sales performance.
3. **The "3-tier rule" origin**: The commonly cited "3-tier" threshold comes from the 2013 joint opinion by SPC, SPP, and MPS (《关于办理组织领导传销活动刑事案件适用法律若干问题的意见》). It establishes the **criminal prosecution standard**: an organization must have **30 or more people AND 3 or more levels** (not counting the organizer's own level) for the organizer/leader to face criminal charges under Criminal Law Article 224-1. **This does NOT mean 3-tier is automatically legal** -- even 2-tier can be administratively penalized if it meets the pyramid selling criteria.
4. **Criminal vs. Administrative distinction**: Criminal Law Article 224-1 only covers "拉人头" and "入门费" type pyramid schemes (with fraudulent intent). "团队计酬" type is generally handled as an administrative violation under the Regulation, unless it is used as a facade for fraud.
5. **Penalties range dramatically**:
- Administrative: 50万-200万元 fine + confiscation of illegal gains (statutory minimum 50万)
- Criminal: up to 15 years imprisonment + fines (the Zhang case had a 1亿元 fine)
- Platform-level: WeChat permanently bans accounts, blocks payment interfaces
6. **2024 enforcement intensity**: In Jan-Sep 2024, procuratorates prosecuted 4,627 people for organizing/leading pyramid selling activities. June 2024 saw SPC+SAMR jointly release 5 typical cyber-pyramid selling cases. October 2024 saw MPS release 5 more typical cases. At least 51 companies were penalized by SAMR in 2024, with 4 cross-jurisdictional enforcement cases.
7. **Major platform cases**:
- **花生日记 (2019)**: 7,456万 fine, up to 51 levels, 20M users. Rectified in 2018.
- **云集 (2017)**: 958万 fine, public account permanently banned. Rectified in 2016, listed on NASDAQ in 2019.
- **永倍达 (2023-2025)**: ~1000亿 RMB involved, 11M users, final conviction in 2025 (Anhui intermediate court), 10 executives sentenced, 5亿 RMB fine (mostly uncollectible).
8. **WeChat anti-pyramid policy**: WeChat's official rules explicitly prohibit multi-level distribution fraud using WeChat relationships. Penalties include feature restrictions to permanent account bans. WeChat Pay merchant review forms specifically check for hierarchical relationships and commission structures above 3 levels. The policy covers public accounts, mini-programs (Article 5.9.3), and video accounts.
9. **Legal distribution model must**: (a) have genuine product transactions at market prices, (b) limit commission tiers to 3 or fewer, (c) base commissions on actual sales profit not recruitment, (d) avoid mandatory entry fees or product purchase requirements, (e) maintain reasonable pricing without inflated "membership" markups.
10. **2026 regulatory trends**: The revision of both the Direct Sales and Anti-Pyramid Selling regulations is in substantive preparation. Potential changes include possible decriminalization/redefinition of "团队计酬" and clearer boundaries for social e-commerce. The regulatory emphasis has shifted from pure prohibition to compliance guidance ("合规深水区").
## Deep Read Notes
### Source 1: SPC/SPP/MPS Joint Opinion on Pyramid Schemes (2013)
This judicial interpretation (via China Law Translate English translation) is the single most important document for understanding the criminal threshold. Key provisions:
- **Article 1**: Defines the criminal threshold as 30+ people and 3+ levels for Criminal Law Article 224-1 prosecution.
- **Article 2**: Clarifies that the "3 levels" count does NOT include the organizer's own level -- meaning 4 total tiers triggers criminal liability.
- **Article 3**: Establishes that "team remuneration" (团队计酬) is only criminal if it is used as a cover for fraud (骗取财物). Legitimate team-based sales compensation is NOT criminal.
- **Article 4**: Defines "organizing and leading" to include founders, decision-makers, key managers, and those who play key roles in recruiting/training.
- **Article 6**: Addresses cross-regional jurisdiction for online pyramid schemes.
This document is the legal basis for the commonly cited "3-tier rule" but importantly distinguishes between administrative illegality and criminal liability.
### Source 3: SPC + SAMR 5 Typical Cases (June 2024)
The joint release by Supreme People's Court and State Administration for Market Regulation on 2024-06-28 covers 5 representative cyber-pyramid selling cases:
1. **Zhang case**: Posed as "poverty alleviation and shared prosperity" charity, illegally obtained 2.5 billion RMB. Principal received 15 years (maximum sentence) + 100 million RMB fine. Demonstrates the "打财断血" (cut off financial resources) principle.
2. **Zhejiang company (Qian et al.) case**: Used "smart charging station mall system" platform to conduct pyramid selling (2020-2022). Demonstrates how new technology/business models are used as cover.
3. **Other cases** in the set involved investment-return schemes, fake public welfare fronts, and new e-commerce models.
Key patterns across all 5 cases: high-return investment promises, fee-based membership qualification, hierarchical organization, recruitment-based compensation. The cases were deliberately selected to show enforcement against diverse new-format pyramid schemes.
### Source 12: 2024 Six Companies Penalized by SAMR
NetEconomy Society (100ec.cn) documented 6 specific administrative penalty cases in 2024:
| # | Company | Penalty | Date |
|---|---------|---------|------|
| 01 | 广州依本生物科技有限公司 | 7万元 fine | 2024-02-21 |
| 02 | 鹤壁市开发区理视康保健服务中心 | 60万余元 confiscated+fined | — |
| 03 | 永乐优选（广东）科技有限公司 | 129万余元 (79万 confiscation + 50万 fine) | 2024-02-28 |
| 04 | 广州麦戈儿科索贸易有限公司 | 50万元 fine | — |
| 05 | 南通社群空间网络科技有限公司 | 120万元 fine | 2024-04-16 |
| 06 | 未来店（广东）网络科技有限公司 | 50万元 fine | 2024-04-22 |
Note that Company #1's 7万元 fine is below the statutory minimum of 50万元, suggesting application of discretionary lighter punishment (possibly voluntary cooperation, small scale, or first offense). The other 5 cases fall within the statutory 50-200万 range.
## Gaps
1. **Full text of the 2026 proposed revision**: The revision is "推动修订" but the draft text has not been publicly released. Cannot assess specific changes to the "团队计酬" definition or tier limits.
2. **WeChat's exact tier enforcement mechanism**: While WeChat's general policy is known, the specific technical implementation of tier monitoring in merchant review forms (《商户业务模式自查报告》) could not be fully accessed.
3. **花生日记 rectification details**: The specific operational changes made by 花生日记 after the 2019 penalty (beyond public statements) are not fully documented in accessible sources.
4. **云集 post-listing compliance model**: How 云集 operates its distribution model post-2019 NASDAQ listing, specifically how it maintains compliance with Chinese law while operating multi-level incentives.
5. **2025-2026 specific enforcement case details**: Many 2025 cases are referenced but full penalty decisions are not publicly accessible, particularly for social e-commerce platforms.
6. **Provincial-level enforcement variation**: Significant variation exists between provinces in enforcement intensity and standards (e.g., Shandong 成武 "趋利执法" controversy in August 2024 where a county-level SAMR allegedly demanded 2100万 from a company). Systematic analysis of this variation is not available.
7. **Direct comparison with US/EU pyramid selling standards**: While the ResearchGate academic paper exists, it is access-restricted. A comparative analysis of China's approach vs. FTC/Kelly Koscot test would strengthen the legal framework understanding.
8. **WeChat video account (视频号) specific rules**: The exact tier limits and enforcement mechanisms for video account-based distribution are not clearly documented.
.claude/skills/deep-research/research-notes/task-f.md
+107
View File
@@ -0,0 +1,107 @@
# Task F: Chinese Social Commerce Distribution Models Research
Research date: 2026-05-06
Researcher: Industry Practice Observer
---
## Sources
| URL | Source-Type | Accessibility | Date | Authority |
|-----|-------------|---------------|------|-----------|
| [拼多多2025年财报](https://www.iyiou.com/news/202603251125182) | 财报/新闻 | Open | 2026-03 | 9 |
| [拼多多2024年总收入](https://www.199it.com/archives/1746028.html) | 数据聚合 | Open | 2025-03 | 8 |
| [Temu GMV 970亿美元](https://xueqiu.com/2226084138/373565004) | 雪球分析 | Open | 2025 | 7 |
| [China Social Commerce Intelligence Report 2025](https://www.globenewswire.com/de/news-release/2025/05/08/3077093/28124/en/China-Social-Commerce-Intelligence-Report-2025-Market-to-Reach-769-Billion-by-2030-Driven-by-Platform-Innovations-Strategic-Partnerships-and-Significant-Investments.html) | 行业报告 | Open | 2025-05 | 8 |
| [抖音电商GMV超30%逼近拼多多](https://finance.sina.com.cn/roll/2025-12-09/doc-inhahnvi2176549.shtml) | 新浪财经 | Open | 2025-12 | 8 |
| [云集被退市警示](https://www.100ec.cn/detail--6604210.html) | 网经社 | Open | 2024-04 | 8 |
| [社交电商大溃败](https://m.cyzone.cn/article/646897) | 创业邦 | Open | 2024-2025 | 8 |
| [社交电商步入合规深水区](https://www.lexology.com/library/detail.aspx?g=6f2dfc03-c164-47d8-8571-3876489b583c) | 法律分析 | Open | 2024 | 9 |
| [花生日记涉传被罚904万](https://m.ebrun.com/ebrungo/zb/422867.html) | 亿邦动力 | Open | 2021-02 | 8 |
| [花生日记官网](https://www.ahsrj.com/) | 企业官网 | Open | 2024 | 7 |
| [城市酷选港股上市](https://hea.china.com/articles/20260421/202604211851437.html) | 中华网 | Open | 2026-04 | 7 |
| [城市酷选冲刺港股估值超10亿](https://news.iresearch.cn/yx/2025/08/531266.shtml) | 艾瑞 | Open | 2025-08 | 8 |
| [城市酷选HKEX招股书](https://www1.hkexnews.hk/listedco/listconews/gem/2025/1211/2025121100660.pdf) | 招股书 | Open | 2025-12 | 9 |
| [城市酷选商业模式合规深度解密](https://m.tech.china.com/redian/2026/0422/042026_1852405.html) | 中华网 | Open | 2026-04 | 6 |
| [有赞分销系统](https://www.youzan.com/intro/fenxiao) | 企业官网 | Open | 2024-2025 | 8 |
| [微三云官网](https://www.weisanyun.cn/) | 企业官网 | Open | 2024-2025 | 7 |
| [HiShop推客分销](https://www.hishop.com.cn/himall/show_161486.html) | 企业官网 | Open | 2025-2026 | 7 |
| [2025中国社交电商行业分析](https://www.chyxx.com/industry/1214588.html) | 产业研究院 | Open | 2025-03 | 8 |
| [小红书猛拉新32元一个](https://www.ebrun.com/20241230/569054.shtml) | 亿邦动力 | Open | 2024-12 | 8 |
| [QuestMobile 2025双十一报告](https://www.questmobile.com.cn/research/report/1991031672729079809/) | 行业报告 | Open | 2025-11 | 9 |
| [社交电商V.S.互联网传销](https://www.chinalawinsight.com/2020/08/articles/intellectual-property/%E7%A4%BE%E4%BA%A4%E7%94%B5%E5%95%86-v-s-%E4%BA%92%E8%81%94%E7%BD%91%E4%BC%A0%E9%94%80/) | 法律分析 | Open | 2020-08 | 9 |
| [有赞2025半年度业绩交流会](https://docs.publicnow.com/viewDoc.aspx?filename=83715%5CEXT%5CD66D47377F05E616A6790B3093B2FCF4F75469C6_2239FC89E1284775BB1FF9190DE96F8930D921EE.PDF) | 业绩纪要 | Open | 2025 | 9 |
---
## Findings
1. **拼多多2025全年营收4318.5亿元（同比+10%）**，增速从2024年的59%显著放缓至10%，标志社交裂变驱动的增长模式已进入存量期。Non-GAAP利润约1025.6亿元。Temu海外GMV从2024年180亿美元跃升至2025年970亿美元，"砍一刀"社交裂变模式成功复制到海外。
2. **云集从传销到退市的完整衰退弧**：2017年因三级分销涉传销被罚958万元后，历经"限制两级分销→会员制电商→私域电商→平台化"多轮转型。但纳斯达克上市4年后市值缩水99%，2024年被退市警示，"会员电商第一股"实质已终结。2024年获国家级高新技术企业称号，但商业模型未能恢复增长。
3. **花生日记罚单从7456万降至904万**：2019年因51级传销链被罚7456万元（社交电商最大罚单），2021年经行政复议撤销原判、改罚904万元。此后完成整改并持续运营至今，官网正常，宣称用户超1亿。但行业热度已大幅下降，未能恢复罚前增长势头。
4. **城市酷选477天港股上市（08050.HK）**：采用"排队免单+异业联盟"模式，宣称覆盖80万+商家、9000万+用户，年交易规模超百亿元。2023年获香港鸿蒙资本1亿元A轮融资，2026年4月港股GEM上市估值超10亿元。2025H1营收约2194万港元、净利1115万港元（扭亏，但主要依赖一次性收入1611万港元）。核心争议：排队免单本质是用新增用户资金为老用户返利，高度依赖持续增长，存在庞氏特征风险。
5. **SaaS平台分销功能矩阵**：
- **有赞**：SaaS年费制，支持无限级代理、多级分销、自动化结算，2024年发布智能化复购分销系统，2025年H1业绩稳定。
- **微三云**：源码买断模式，提供上百种分销模式（链动2+1、排队补贴、众店模式等），适合技术团队二次开发。
- **HiShop**：老牌系统商，2025年主推"推客分销"、微信小店深度适配、全场景带货模式，2026年趋势为合规化运营+按效果付费。
6. **2024-2026行业趋势**：中国社交电商GMV从2023年3.42万亿元增长，预计2030年达7690亿美元。抖音电商2024年GMV约3.5万亿元、2025年目标4.2-4.3万亿元（增速超30%逼近拼多多）。社交电商占网络零售比例预计2025年达17.1%。核心趋势：从"流量裂变"转向"心智占领"、直播电商主导、AI驱动精细化运营、合规化高压。
7. **CAC对比**：拼多多早期社交裂变获客成本仅5-7元/人（远低于行业），但2025年CAC持续攀升。小红书2024年地推拉新约32元/人。抖音广告CPE约2.0元。全行业2024年平均获客成本同比上涨约23%。电商已进入存量竞争时代。
8. **快手差异化优势**：用户月均复购率4.2次（抖音1.5次、拼多多3.1次），深耕"熟人社交"在下沉市场壁垒高。2025年四大平台（抖音、拼多多、快手、小红书）年交易额合计超8万亿元。
9. **"三级分销"法律红线持续收紧**：2024-2025年社交电商合规监管高压，Lexology等法律分析指出"入门费+拉人头+下线计酬"三要素同时出现即构成传销。各SaaS平台已将分销层级压缩至两级或以下以规避风险。
10. **拼多多"千亿扶持"计划**：2025年4月正式推出，未来三年投入超1000亿元构建用户-商家-平台共赢生态，标志从"低价掠夺"转向"生态共建"的战略调整。
---
## Deep Read Notes
### Source 1: 城市酷选HKEX招股书 (2025-12)
- **类型**：官方招股书，权威性最高
- **关键数据**：聚合约90万家商户；收益2193.7万港元（2025H1）；扭亏为盈依赖一次性收入（1611万港元），核心业务盈利能力存疑
- **模式分析**：招股书将平台定位为"SaaS软件服务商"而非交易平台运营方，强调"不发起、不参与、不运营"线下排队免单活动——这种法律定位设计明显是为了规避传销/非法集资的监管风险
- **风险信号**：排队免单模式的资金池运作透明度不足，用户提现规则存在争议；核心增长驱动力高度依赖新增商户和消费者的持续涌入
### Source 2: 云集退市警示与社交电商大溃败 (创业邦/网经社, 2024-2025)
- **类型**：行业深度分析
- **关键叙事**：云集的衰退代表了整个"社交电商"品类的系统性困境——从2019年的资本狂热到2024年的"集体幻灭"
- **核心洞察**：
1. 云集转型路径（三级分销→两级→会员制→私域→平台化）本质上是在不断剥离增长引擎以换取合规性，每"整改"一次就失去一层增长动力
2. 贝店（云集旗下）2021年爆雷是标志性事件——社交电商"拉人头"模式已被证明在监管高压下不可持续
3. 到2025年，真正存活的"社交电商"已从独立平台转型为依附于微信/抖音生态的分销工具（SaaS化），而非独立流量入口
- **对研究的启示**：社交电商的"分销"概念已从"多层级人头费"演变为"内容驱动的推荐佣金"——这是合规化后的根本范式转移
### Source 3: 拼多多2025财报 + Temu增长分析
- **类型**：财报数据 + 分析师解读
- **关键数据**：2025全年营收4318.5亿元（+10%），增速从2024年59%断崖式下降；Temu GMV从180亿→970亿美元
- **深层分析**：
1. 国内拼多多"拼团裂变"模式的增长天花板已现——7.2亿年活买家接近中国网购人口上限，CAC持续攀升
2. 增长接力棒已交给Temu——"砍一刀"社交裂变在北美/欧洲同样有效，但面临地缘政治和监管风险
3. "千亿扶持"计划（1000亿元/3年）是战略转向信号：从"零和博弈"（压榨商家）转向"正和博弈"（生态共建），但短期内可能进一步压缩利润率
---
## Gaps
1. **城市酷选完整财务数据**：HKEX招股书仅披露了部分中期数据，年度营收、GMV具体口径（是否含退款/刷单）无法核实。招股书中"排队免单"资金池的会计处理方式不明确。
2. **微三云实际客户数据**：作为源码出售模式，缺乏公开的客户数量、续费率、GMV贡献等核心经营指标。
3. **花生日记2024-2025运营数据**：仅有"用户超1亿"的官方宣称，缺乏第三方验证。APP下载量和活跃度数据缺失。
4. **SaaS平台分销功能合规性对比**：各平台如何将分销层级控制在法律红线（三级）以内的具体技术方案缺乏系统性对比。
5. **CAC精确对比数据**：各平台的获客成本数据口径不一（拼多多用营销费用率推算、小红书用地推成本、抖音用广告CPE），无法做标准化横向对比。
6. **"排队免单"模式法律定性**：目前缺乏权威司法判例明确判定"排队免单"是否构成传销或非法集资，城市酷选的合规辩护是否经得起法律检验尚无结论。
7. **2026年最新政策变化**：社交电商/分销模式的最新监管政策（特别是2025年底至2026年初的法规更新）搜索覆盖不足。
8. **快手电商GMV和利润数据**：在2024-2025年社交电商格局中的具体GMV、盈利状况缺乏详细数据支撑。