Gingiris commited on
Commit
cf243e6
·
verified ·
1 Parent(s): 7a01eca

Add PLG signal gates and evidence standards

Browse files
Files changed (1) hide show
  1. SKILL.md +280 -64
SKILL.md CHANGED
@@ -1,95 +1,311 @@
1
  ---
2
  name: gingiris-b2b-growth
3
  description: |
4
- 🇺🇸 B2B SaaS Full-Lifecycle Growth Playbook — From PMF validation to ecosystem growth. Real cases from HeyGen, Deel, Vercel, Supabase, AWS. PLG/SLG strategies, affiliate marketing, channel partnerships.
5
-
6
- 🇨🇳 AI 全球 B2B 产品增长指南 PMF 验证到生态化增长的完整操作手册。整合 HeyGenDeel、Vercel、Supabase、AWS 等标杆案例。PLG/SLG 策略、联盟营销、渠道合作
7
-
8
- 🇯🇵 B2B SaaS成長プレイブック — PMF検証からエコシステム成長まで。HeyGen、Deel、Vercel、Supabase、AWSの実例。PLG/SLG戦略、アフィリエイト、チャネルパートナーシップ。
9
-
10
- 🇰🇷 B2B SaaS 성장 플레이북 — PMF 검증부터 생태계 성장까지. HeyGen, Deel, Vercel, Supabase, AWS 실제 사례. PLG/SLG 전략, 제휴 마케팅, 채널 파트너십.
11
-
12
- Triggers: "B2B growth" | "PMF" | "product-market fit" | "PLG" | "SLG" | "affiliate marketing" | "channel partners" | "customer case study" | "B2B 增长" | "用户调研"
13
- ---
14
 
15
- # AI 全球 2B 产品全生命周期增长指南
16
 
17
- > 🌍 **Language / 语言**: [中文](#中文版) | [English](https://github.com/Gingiris-1031/gingiris-b2b-growth/blob/main/references/en/README.md) | [日本語](https://github.com/Gingiris-1031/gingiris-b2b-growth/blob/main/references/ja/README.md) | [한국어](https://github.com/Gingiris-1031/gingiris-b2b-growth/blob/main/references/ko/README.md)
18
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
19
  ---
20
 
21
- ## 中文版
 
 
 
 
22
 
23
- > 版本:v4.0(终极融合版)| 整合 HeyGen、Deel、Vercel、Supabase、AWS 等标杆案例
 
 
24
 
25
- ### 核心框架:增长飞轮
26
 
27
  ```
28
- ┌─────────────┐
29
- │ 产品体验 │ ← PLG 核心
30
- └──────┬──────┘
31
-
32
- ┌──────▼──────┐
33
- │ 社区运营 │ ← CLG 放大
34
- └──────┬──────┘
35
-
36
- ┌──────▼──────┐
37
- │ 渠道生态 │ ← 规模化
38
- └──────┬──────┘
39
-
40
- ┌──────▼──────┐
41
- │ 直销转化 │ ← SLG 变现
42
- └─────────────┘
43
  ```
44
 
45
- ### 详细指南(中文)
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
46
 
47
- | 主题 | 文件 |
 
 
48
  |:---|:---|
49
- | 奠定基石(PMF、用户调研、社区) | [references/foundation.md](https://github.com/Gingiris-1031/gingiris-b2b-growth/blob/main/references/foundation.md) |
50
- | 构建增长引擎(PLG/SLG、联盟、渠道) | [references/growth-engine.md](https://github.com/Gingiris-1031/gingiris-b2b-growth/blob/main/references/growth-engine.md) |
51
- | 价值呈现(销售Deck、案例叙事) | [references/value-presentation.md](https://github.com/Gingiris-1031/gingiris-b2b-growth/blob/main/references/value-presentation.md) |
52
- | 合同模板体系 | [references/contracts.md](https://github.com/Gingiris-1031/gingiris-b2b-growth/blob/main/references/contracts.md) |
53
- | 数据驱动增长 | [references/metrics.md](https://github.com/Gingiris-1031/gingiris-b2b-growth/blob/main/references/metrics.md) |
54
 
55
- ---
 
 
 
 
56
 
57
- ## Quick Navigation (All Languages)
 
 
58
 
59
- | 🇨🇳 中文 | 🇺🇸 English | 🇯🇵 日本語 | 🇰🇷 한국어 |
60
- |:---|:---|:---|:---|
61
- | [奠定基石](https://github.com/Gingiris-1031/gingiris-b2b-growth/blob/main/references/foundation.md) | [Foundation](https://github.com/Gingiris-1031/gingiris-b2b-growth/blob/main/references/en/foundation.md) | [基盤構築](https://github.com/Gingiris-1031/gingiris-b2b-growth/blob/main/references/ja/foundation.md) | [기반 구축](https://github.com/Gingiris-1031/gingiris-b2b-growth/blob/main/references/ko/foundation.md) |
62
- | [增长引擎](https://github.com/Gingiris-1031/gingiris-b2b-growth/blob/main/references/growth-engine.md) | [Growth Engine](https://github.com/Gingiris-1031/gingiris-b2b-growth/blob/main/references/en/growth-engine.md) | [成長エンジン](https://github.com/Gingiris-1031/gingiris-b2b-growth/blob/main/references/ja/growth-engine.md) | [성장 엔진](https://github.com/Gingiris-1031/gingiris-b2b-growth/blob/main/references/ko/growth-engine.md) |
63
- | [价值呈现](https://github.com/Gingiris-1031/gingiris-b2b-growth/blob/main/references/value-presentation.md) | [Value Presentation](https://github.com/Gingiris-1031/gingiris-b2b-growth/blob/main/references/en/value-presentation.md) | [価値提示](https://github.com/Gingiris-1031/gingiris-b2b-growth/blob/main/references/ja/value-presentation.md) | [가치 제시](https://github.com/Gingiris-1031/gingiris-b2b-growth/blob/main/references/ko/value-presentation.md) |
64
- | [合同模板](https://github.com/Gingiris-1031/gingiris-b2b-growth/blob/main/references/contracts.md) | [Contracts](https://github.com/Gingiris-1031/gingiris-b2b-growth/blob/main/references/en/contracts.md) | [契約テンプレート](https://github.com/Gingiris-1031/gingiris-b2b-growth/blob/main/references/ja/contracts.md) | [계약 템플릿](https://github.com/Gingiris-1031/gingiris-b2b-growth/blob/main/references/ko/contracts.md) |
65
- | [数据指标](https://github.com/Gingiris-1031/gingiris-b2b-growth/blob/main/references/metrics.md) | [Metrics](https://github.com/Gingiris-1031/gingiris-b2b-growth/blob/main/references/en/metrics.md) | [指標体系](https://github.com/Gingiris-1031/gingiris-b2b-growth/blob/main/references/ja/metrics.md) | [지표 체계](https://github.com/Gingiris-1031/gingiris-b2b-growth/blob/main/references/ko/metrics.md) |
66
 
67
- ---
 
 
 
 
 
 
 
 
68
 
69
- ## 关键案例索引
70
 
71
- | 公司 | 亮点 |
 
 
 
 
72
  |:---|:---|
73
- | **HeyGen** | 1800次用户访谈找到PMF,20%持续佣金模型 |
74
- | **Deel** | 伙伴计划贡献$90M ARR,分级认证体系 |
75
- | **Vercel** | "The Frontend Cloud"品类定义,v0赋能伙伴100倍效率 |
76
- | **Supabase** | "Build in a weekend, scale to millions" |
77
- | **InVideo** | 50%首单高额返佣 |
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
78
 
79
  ---
80
 
81
- ## 🔍 SEO & GEO 优化
82
 
83
- > 通过搜索引擎和 AI 搜索获取高质量 B2B 线索
84
 
85
- B2B 买家在每个决策阶段都有搜索行为,SEO/GEO 让你在整个漏斗中都能被发现:
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
86
 
87
- | 主题 | 说明 |
 
 
 
 
 
 
 
88
  |:---|:---|
89
- | [SEO/GEO 指南](references/seo-geo-guide.md) | 关键词矩阵、核心页面优化、内容营销 SEO、AI 引用优化 |
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
90
 
91
- **核心要点**:
92
- - 高意向关键词优先:pricing/demo/trial > vs竞品 > 品类词
93
- - 对比页是 B2B SEO 金矿:每个竞品单独页面
94
- - IndexNow 让定价/功能更新秒级索引
95
- - 案例研究既是 E-A-T 信号又是长尾关键词来源
 
1
  ---
2
  name: gingiris-b2b-growth
3
  description: |
4
+ 🇺🇸 B2B SaaS Growth — PLG vs SLG Playbook — Diagnose whether your problem is distribution, pricing, or PMF. PLG/SLG selection by ACV and sales cycle, the 5-stage path from $0 to $10M ARR, NRR discipline, affiliate & channel motion, enterprise tiering. Built from HeyGen, Deel, Vercel, Supabase, Snowflake patterns.
5
+
6
+ 🇨🇳 B2B SaaS 增长 — PLG vs SLG 决策手册 — 诊断你的瓶颈是分发定价还是 PMF按 ACV 和销售周期选 PLG/SLG,$0 → $10M ARR 的 5 阶段路径,NRR 纪律,联盟渠道打法,Enterprise 分层提炼自 HeyGen、Deel、Vercel、Supabase、Snowflake。
 
 
 
 
 
 
 
7
 
8
+ 🇯🇵 B2B SaaS グロース — PLG vs SLG プレイブック — 課題が流通・価格・PMFのどれかを診断。ACVとセールスサイクルでPLG/SLG選定、$0→$10M ARRの5段階パス、NRR規律、アフィリエイト&チャネル。
9
 
10
+ 🇰🇷 B2B SaaS 성장 PLG vs SLG 플레이북 — 병목이 유통·가격·PMF 중 무엇인지 진단. ACV와 세일즈 사이클로 PLG/SLG 선택, $0 → $10M ARR 5단계 경로, NRR 규율, 제휴 & 채널.
11
 
12
+ Triggers: "B2B growth" | "B2B SaaS" | "PLG" | "SLG" | "product-led growth" | "sales-led growth" | "PLG or SLG" | "SaaS pricing" | "ARR growth" | "NRR" | "go to market" | "enterprise sales" | "affiliate" | "PMF" | "B2B 增长" | "PLG 还是 SLG" | "SaaS 定价" | "渠道合作" | "PLG/SLG 선택"
13
+ when_to_use: |
14
+ Use this skill when you need to: diagnose PLG vs SLG motion, grow B2B SaaS revenue from
15
+ $0 to $10M ARR, design SaaS pricing tiers, build affiliate/channel programs, validate PMF,
16
+ improve NRR, or structure enterprise sales motion.
17
+ Trigger phrases: "B2B growth" | "PLG vs SLG" | "SaaS pricing" | "enterprise sales" |
18
+ "ARR growth" | "B2B增长" | "PLG还是SLG" | "SaaS定价" | "渠道合作"
19
+ source: https://github.com/Gingiris-1031/gingiris-skills/tree/main/skills/gingiris-b2b-growth
20
+ tags:
21
+ - b2b-growth
22
+ - plg
23
+ - slg
24
+ - saas
25
+ - saas-marketing
26
+ - product-led-growth
27
+ - sales-led-growth
28
+ - b2b-saas
29
+ - startup-growth
30
+ - go-to-market
31
+ - revenue-growth
32
+ - mrr
33
+ - claude-code
34
+ - ai-agent-skill
35
+ - latest
36
  ---
37
 
38
+ # B2B SaaS Growth — PLG / SLG Playbook
39
+
40
+ > 🌍 **Language / 语言**: [English](#english-version) | [中文](#中文版)
41
+
42
+ "People like it but revenue isn't growing" is rarely one problem. It's usually a **mis-matched motion**: a PLG product run with an SLG playbook, or vice versa. Start by diagnosing the motion, then walk the staged path.
43
 
44
+ ---
45
+
46
+ ## English Version
47
 
48
+ ### Pick your motion (the flywheel)
49
 
50
  ```
51
+ PLG (product-driven)
52
+
53
+ existing traffic + short decision
54
+ ←─────────────────────────────────→
55
+ needs trust + long cycle
56
+
57
+ SLG (sales-driven)
 
 
 
 
 
 
 
 
58
  ```
59
 
60
+ | Signal | Motion | Reference |
61
+ |:---|:---|:---|
62
+ | ACV < $1k/yr **and** decision < 1 week | **PLG** | Vercel, Supabase |
63
+ | ACV > $10k/yr **and** decision > 1 month | **SLG** | Snowflake, Databricks |
64
+ | In between | **PLG-to-SLG hybrid** | HeyGen, Deel |
65
+
66
+ ### The 5-stage path
67
+
68
+ | Stage | ARR | Key move |
69
+ |:---|:---|:---|
70
+ | **Pre-PMF** | $0–$100k | 10–30 user interviews; confirm the must-have before anything else |
71
+ | **Early PMF** | $100k–$1M | Invest in **1–2 channels deeply** — do not spray |
72
+ | **Growth** | $1M–$5M | Add affiliate + integration partners; PLG gets sales-assist |
73
+ | **Scale** | $5M–$10M | Enterprise team, compliance (SOC 2), channel acceleration |
74
+ | **Ecosystem** | $10M+ | Open platform, investor / agency distribution |
75
+
76
+ ### Conversion-first: grind signup-to-paid before you scale
77
 
78
+ Counter-intuitive but field-proven: do not scale traffic until signup-to-paid conversion clears the bar. Below it, paid traffic is negative-ROI ("suicide marketing").
79
+
80
+ | Benchmark | Threshold |
81
  |:---|:---|
82
+ | ToC AI signup-to-paid | **>= 1.5%** before scaling (good products: 6-8%; best seen ~40%) |
83
+ | ToB AI signup-to-paid | **> 10-15%** before scaling |
84
+ | Teams that actually executed this | 4-8%+ signup-to-paid, and smoother fundraising |
 
 
85
 
86
+ The SOP (200+ teams heard it; fewer than 10 executed):
87
+ 1. Freeze paid ads, influencer pushes, and launch campaigns until the benchmark is hit.
88
+ 2. Find the persona with the highest willingness to pay - talk to one user 5 times if needed.
89
+ 3. Resist FOMO from competitors' PR; hold the line for months.
90
+ 4. Proof case: one coached ToC product stayed quiet for ~6 months, reached 6-8% signup-to-paid - ~$5,000 MRR on ~10,000 registered users - then scaled, and the growth curve (and fundraising) took off.
91
 
92
+ Unit economics, both ends:
93
+ - Interview-driven path: Gamma, HeyGen, Lovable each ran ~500-1,000 user interviews within roughly half a year before scaling (self-reported industry accounts). HeyGen's 5-minute signup-to-pay flow was distilled from ~1,800 interviews - 1 free credit, paywall exactly at the aha moment.
94
+ - Burn path (anonymized counterexample): tens of dollars per signup, hundreds of dollars per paying user.
95
 
96
+ ### Day-0 pricing discipline
 
 
 
 
 
 
97
 
98
+ - **Charge from Day 0.** A pricing-test window is fine, but collect money - even $1 plants the "this is not a free product" perception. Paying users vote with their wallet; free users just have opinions.
99
+ - **Don't wait for the paid tier to exist**: ship a (deliberately vague) pricing page with a contact form. Field result: dozens of price inquiries within two weeks; inquiry mix ~70% community-free / 20% personal / 10% team - and the question shifts from "is this free?" to "how much?"
100
+ - Users genuinely pay for **at most ~3 points** in any product. Instrument usage, find them, gate there.
101
+
102
+ ### PLG-to-SLG watershed and lean-team benchmarks
103
+
104
+ - Below **~$1M ARR**: PLG + organic + light paid marketing is enough. Above ~$1M ARR (and most ToB): you need SLG / local key-account channels to keep scaling.
105
+ - Lean-team bar (podcast-recalled figures, treat as approximate): Notion ~4-8 people at million-user scale; ClickUp ~24 people at $1M ARR; several 8-10 person teams in the same space at $1M revenue. Operator rule: **over 25 people without $1M revenue = the org is the problem, not the market.**
106
+ - Inbound-heavy SLG pattern (founder's own account, self-reported): Mattermost's community edition drives search ranking, self-serve trials, then inbound deals - growing from ~$100k contracts to ~$1M contracts without a large sales team. Dify sells a standardized on-prem edition at ~$30-50k per deal with a single deploy engineer, routing custom work to partner vendors.
107
 
108
+ ### Enterprise deal execution (private deployment & custom 2B)
109
 
110
+ Winning the contract is half the job — getting paid in full is the other half. Distilled from a Multi-Agent open-source commercialization project retro.
111
+
112
+ #### Payment structure & final-installment defense
113
+
114
+ | Rule | Detail |
115
  |:---|:---|
116
+ | Staged payments | Private deployment splits **3:4:3** or **4:4:2** — first features live / full deployment / final payment after 3–6 months of ops |
117
+ | Collect on signature | **30% due the day the contract is signed** |
118
+ | Late penalty | Written into the contract: **0.5% per day** overdue |
119
+ | Final-payment leverage | Keep 1–2 features deployed on **your own servers** until the last payment clears — say it before signing, write it into the terms |
120
+ | Who to chase | Chase the tail payment with the **highest-ranking contact**; if your champion stalls, escalate to the CEO once or twice |
121
+ | Long-term direction | Every new deal: prepayment up, tail payment down |
122
+
123
+ #### Scope control: SOR / SOW / CCB
124
+
125
+ - Keep **SOR** (customer's raw asks) separate from **SOW** (acceptance criteria + delivery milestones + change terms); **sign the SOW together with the contract**.
126
+ - All changes go through a **CCB**: customer initiates in writing → both PMs assess impact (cost / schedule) → both sides sign a change confirmation.
127
+ - Leave a paper trail for everything; name the **acceptance contact** in the contract.
128
+ - Customer gets excited on a call and promises the moon? **Only paperwork counts.**
129
+
130
+ #### Commercial push-pull
131
+
132
+ - Customer asks for 50 features and you could do 40 → "we can commit to 10, and because we want this to succeed we'll stretch to 15." Turn concessions into favors.
133
+ - Requests you don't want → **price them up** (pad the person-days) instead of refusing outright.
134
+ - Private deployment has only **two legitimate drivers**: data privacy + internal permission-system integration. Everything else belongs on SaaS.
135
+ - SaaS customization → sell a **License**. SSO and similar integrations → open API + standard docs, let the customer build it themselves.
136
+
137
+ #### Revenue & channel math
138
+
139
+ | Path to $1M ARR | Breakdown |
140
+ |:---|:---|
141
+ | Enterprise route | 9 × $100k 2B customers |
142
+ | Subscription route | ~~5,000 × $200/yr subscribers (≈500 per $100k) per $100k (≈5,000 for $1M) |
143
+
144
+ | Channel tier | Terms |
145
+ |:---|:---|
146
+ | Reseller | 10–20% referral fee |
147
+ | Silver (>$100k/yr sales) | 15% |
148
+ | Gold (>$300k/yr sales) | 20% |
149
+
150
+ Reseller due diligence — three questions: what did you resell before? where do your customers come from? reseller or service partner?
151
+
152
+ **2B content formula**: 80% how you helped customers succeed + 20% about the customer. Deck narrative: pain first, then the fix.
153
+
154
+ > Deep dive: `references/deal-execution.md`. The full operator layer (contract-clause quick cards, the 12-question pre-signing checklist, negotiation scripts) ships with the Gingiris paid field manual → https://gingiris.tools/playbook/2b-deals/
155
+
156
+ ### B2B content is different from B2C
157
+
158
+ - **Must** have a case study (ROI / time saved / headcount saved).
159
+ - **Must** have verifiable numbers ("customer X saved Y hours/month").
160
+ - **Must** have a CFO-friendly pricing page (ROI calculator, annual discount, compliance docs).
161
+ - ❌ No B2C hype words ("amazing", "magical").
162
+
163
+ ### Anti-patterns
164
+
165
+ - ❌ Paid acquisition before PMF — you burn cash 10× faster than you learn.
166
+ - ❌ Forcing PLG → SLG at $1M just because. Vercel held PLG to $100M.
167
+ - ❌ Watching ARR but ignoring **NRR**. NRR < 100% = a leaking bucket.
168
+ - ❌ One contract template for everyone — split startup / mid-market / enterprise.
169
+
170
+ ### What to do next
171
+
172
+ - Pre-PMF → run formal user interviews to find the must-have.
173
+ - Post-PMF, need content → move to the blog/SEO workflow (B2B voice: data > story).
174
+ - Going global → localize the listing and pricing.
175
+ - Benchmarking a rival → run a competitor teardown.
176
 
177
  ---
178
 
179
+ ## 中文版
180
 
181
+ ### 选你的增长引擎(飞轮)
182
 
183
+ ```
184
+ PLG(产品驱动)
185
+
186
+ 已有流量 + 短决策
187
+ ←─────────────────────────────────→
188
+ 需要信任 + 长周期
189
+
190
+ SLG(销售驱动)
191
+ ```
192
+
193
+ | 信号 | 引擎 | 对标 |
194
+ |:---|:---|:---|
195
+ | ACV < $1k/年 **且** 决策 < 1 周 | **PLG** | Vercel、Supabase |
196
+ | ACV > $10k/年 **且** 决策 > 1 个月 | **SLG** | Snowflake、Databricks |
197
+ | 中间 | **PLG-to-SLG 混合** | HeyGen、Deel |
198
+
199
+ ### 5 阶段路径
200
+
201
+ | 阶段 | ARR | 关键动作 |
202
+ |:---|:---|:---|
203
+ | **Pre-PMF** | $0–$100k | 10–30 次用户访谈,先确认 must-have |
204
+ | **Early PMF** | $100k–$1M | **投 1–2 个渠道做深**,不要广撒网 |
205
+ | **Growth** | $1M–$5M | 引入 affiliate + 集成伙伴,PLG 加 sales-assist |
206
+ | **Scale** | $5M–$10M | Enterprise 团队、合规(SOC 2)、渠道加码 |
207
+ | **Ecosystem** | $10M+ | 开放平台、investor / agency 分销 |
208
+
209
+ ### 转化优先:先磨注册→付费,再放量
210
+
211
+ 反常识但实战验证:注册→付费转化率没到基准线之前,不要放量——投放 ROI 打不正("自杀式营销")。
212
+
213
+ | Benchmark | 基准线 |
214
+ |:---|:---|
215
+ | ToC AI 注册→付费 | **≥1.5%** 才可放量(好产品 6–8%,最好见过 ~40%) |
216
+ | ToB AI 注册→付费 | **>10–15%** 才可放量 |
217
+ | 真正执行者的结果 | 注册→付费 4–8%+,融资也更顺 |
218
+
219
+ SOP(200+ 团队听过,真正执行完的不到 10 个):
220
+ 1. 达标前冻结投放、红人、launch campaign。
221
+ 2. 找出当前阶段付费意愿最高的画像——必要时"和一个用户聊 5 次"。
222
+ 3. 抵抗竞品 PR 带来的 FOMO,顶住几个月。
223
+ 4. 实证:一个被辅导的 ToC 产品沉寂约半年,注册→付费做到 6–8%——约 1 万注册用户做到 ~$5,000 MRR——之后放量曲线起飞、融资顺利。
224
+
225
+ 单位经济学两端对照:
226
+ - 古典产品主义:Gamma、HeyGen、Lovable 都在约半年内做了 ~500–1,000 场用户访谈再放量(口述行业参照)。HeyGen 的"注册到付费 5 分钟"链路提炼自 ~1,800 场访谈——只给 1 个 free credit,付费墙卡在 aha moment 最高点。
227
+ - 烧钱投放反例(匿名):单个注册成本几十美金、单个付费用户成本几百美金。
228
+
229
+ ### Day-0 收费纪律
230
+
231
+ - **从 day 0 开始收钱。** 允许 pricing test 调整期,但要勇敢收费——哪怕收 $1,也种下"这不是免费产品"的心智。付费用户用脚投票,免费用户只是发表意见。
232
+ - **不要等付费版做出来**:先挂一个(刻意写模糊的)pricing 页 + contact 表单。实测:两周收到几十个问价;比例约 70% community free / 20% 个人 / 10% team——用户口径从"这免费吗"变成"大概多少钱"。
233
+ - 一个产品用户真正付费的点**不超过 3 个**。做埋点找到它们,把卡点设在那里。
234
+
235
+ ### PLG→SLG 分水岭与精简人效基准
236
+
237
+ - **~$1M ARR 以下**:PLG + organic + 常规投放足够;~$1M ARR 以上(及多数 ToB):必须配 SLG / 本地 KA 渠道才能持续规模化。
238
+ - 人效基准(播客口述回忆,按约数看):Notion 约 4–8 人时已达百万用户量级;ClickUp 约 24 人做到 $1M ARR;同赛道多个 8–10 人团队做到百万美金营收。操盘铁律:**超过 25 人还没到 $1M 营收 = 组织有问题,不是市场有问题。**
239
+ - Inbound 型 SLG 范式(founder 面授口述):Mattermost 社区版带搜索排名→用户自助试用→主动 inbound,从 ~$10 万一单做到 ~$100 万一单,无需养大型销售团队。Dify 只卖标准化私有部署版(~$3–5 万��金/单,派一个工程师部署),定制需求全部转给合作 vendor。
240
+
241
+ ### 2B 打单与交付(私有化 & 定制)
242
+
243
+ 签下合同只是一半,全额收回款才是另一半。提炼自某 Multi-Agent 开源商业化项目复盘。
244
+
245
+ #### 付款结构与尾款防御
246
+
247
+ | 规则 | 细节 |
248
+ |:---|:---|
249
+ | 分段付款 | 私有化部署 **3:4:3** 或 **4:4:2** —— 第一批功能上线 / 全部部署完 / 运维 3–6 个月后付尾款 |
250
+ | 签约即收款 | **签合同当天收 30%** |
251
+ | 违约金 | 直接写进合同:**拖一天多付 0.5%** |
252
+ | 尾款筹码 | 交付时留 1–2 个功能部署在**自己服务器**,尾款结清才迁 —— 签合同前说清楚、条款里写清楚 |
253
+ | 找谁要 | 尾款直接找**对接的最高的人**;对接人推诿可问 1–2 次 CEO |
254
+ | 长期方向 | 每一单:预付款越来越高、尾款越来越少 |
255
+
256
+ #### 需求边界管理:SOR / SOW / CCB
257
+
258
+ - **SOR**(客户原始输入)与 **SOW**(验收标准 + 交付节点 + 变更条款)分开;**合同和 SOW 一起签**。
259
+ - 变更一律走 **CCB**:客户先书面发起 → 双方项目经理评估影响(金额 / 工期)→ 双方签变更确认文件。
260
+ - 一切留痕(会议 / 对话 / 验收记录);合同指明"项目验收对接人"。
261
+ - 客户聊嗨了拍胸脯?**只有 paperwork 有效。**
262
+
263
+ #### 商务推拉
264
+
265
+ - 客户提 50 个需求、内部判断 40 能做 →"先答应 10 个,因为想帮你所以做到 15 个"——把让步做成人情。
266
+ - 不想做的需求**抬价过滤**(人/天多标几天),不用直接拒绝。
267
+ - 私有化需求只有**两个真点**:数据隐私 + 接内部权限系统,其他都该买 SaaS。
268
+ - SaaS 二开 → 直接卖 **License**;SSO 等集成 → 开放 API + 标准手册,让客户自己做。
269
 
270
+ #### 营收与渠道账
271
+
272
+ | $1M ARR 拆解 | 组成 |
273
+ |:---|:---|
274
+ | 2B 路线 | 9 个 $10 万企业客户 |
275
+ | 订阅路线 | 每 $10 万 ≈ 500 个 $200/年订阅用户($1M ≈ 5,000) |
276
+
277
+ | 代理分层 | 条件 |
278
  |:---|:---|
279
+ | Reseller | 10–20% 推荐费 |
280
+ | 银牌(年销 >$10 万)| 15% |
281
+ | 金牌(年销 >$30 万)| 20% |
282
+
283
+ 代理尽调三问:之前代理过什么产品?客户从哪来?Reseller 还是 Service Partner?
284
+
285
+ **2B 内容公式**:80% 讲自己如何帮客户成功 + 20% 讲客户;deck 叙事"先讲怎么痛,再说怎么好"。
286
+
287
+ ### B2B 内容特殊要求
288
+
289
+ - **必须有 case study**(ROI / 省时 / 省人力)。
290
+ - **必须有可验证数据**("customer X saved Y hours/month")。
291
+ - **必须有 CFO 友好的 pricing 页**(ROI 计算器、年付折扣、合规文档)。
292
+ - ❌ 不用 ToC 话术("amazing"、"magical")。
293
+
294
+ ### 反模式
295
+
296
+ ### PLG 信号门(2026 会议纪要更新)
297
+
298
+ - 免费 DAU 不等于 PLG 成立。先定义一个与价值直接相关的激活事件,再看 `注册 → 激活 → 重复使用 → 付费/合格扩张`。
299
+ - 剔除内部、测试和活动赠送用户,并按注册月份做 cohort;只有登录、没有完成核心任务的用户不算激活。
300
+ - 产品未达到稳定激活与留存前,不扩付费投放、不盲目加销售团队;回到访谈、onboarding 和定价实验。
301
+ - 当团队使用、席位增长、安全/采购需求或明确扩张意图出现时再引入 sales-assist,而不是见到免费用户就交给销售。
302
+ - 所有转化阈值注明日期、样本范围与数据来源;历史案例用于形成假设,不作为跨产品的永久标准。
303
+
304
+ - ❌ 没 PMF 就投流(烧钱比学习快 10x)。
305
+ - ❌ 到 $1M 就盲目 PLG → SLG(Vercel 坚持 PLG 到 $100M)。
306
+ - ❌ 只看 ARR 不看 **NRR**(NRR < 100% = 漏水桶)。
307
+ - ❌ 只给一个合同模板(分 startup / mid-market / enterprise 三版)。
308
+
309
+ ---
310
 
311
+ *By Iris (生姜 Iris) · ex-COO @ AFFiNE, 150+ AI startups advised · Install: `npx skills add Gingiris-1031/gingiris-b2b-growth`*