network-effects
Skill1 starsChanged 8 months agoMITView on GitHub
Andrew Chen's Cold Start Problem framework — atomic network definition with thresholds by product type, anti-peanut-buttering, zero tracking, escape velocity 3-force decomposition (engagement/acquisition/economic), growth ceiling 5-force detection, competitive position dynamics, T2D3 benchmark. Use for products with network effects, marketplace dynamics, or viral growth potential.
We split this skill into the rules it gives an agent
2 rules. About 0 are common, 2 are distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈2
Distinctive · estimate
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 12
When evaluating viral growth potential
- Distinctive: line 50
No listings for a search
≈ “Common” is an estimate: at least 1 in 5 of the 308,232 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Network Effects Strategy
When to Apply
- Product's value increases with more users (network-effect products)
- Marketplace, social, or community products
- When evaluating viral growth potential
- Cold start problems (chicken-and-egg)
- Growth ceiling diagnosis
Core Framework
Atomic Network Definition
The smallest group where the product becomes self-sustaining. Define during onboarding:
"Does your product's value increase as more people use it?"
-> YES:
"Let's define your Atomic Network — the smallest self-sustaining group.
Reference thresholds:
Slack: 3 people + 2,000 messages = 93% long-term retention
Zoom: 2 people (smallest possible)
Airbnb: 300 listings + 100 reviewed per city
Uber: 15-20 concurrent drivers, ETA < 3 min per city
Tinder: 20,000 users in a single market
What is the minimum unit for YOUR product?
How many users in one [team/city/community] make it work?"Finding the threshold: Plot network size (X) vs engagement metric (Y). Look for the inflection point ("kink") where engagement spikes. Everything before the kink is anti-network-effect territory.
Anti-Peanut-Buttering
"Do NOT spread across segments until atomic network is achieved in the first one. Density > breadth."
Concentrating resources in one segment to achieve critical mass beats thinly spreading across many.
"Zero" Tracking
A "zero" = the product's core value cannot be fulfilled because the network is too sparse:
- No drivers available
- No listings for a search
- AI gives generic recommendation
Track zero rates per segment. Users who get "zeroed" churn AND don't return — the damage persists.
Escape Velocity — 3-Force Decomposition
Once atomic network is established, track three independent growth forces:
| Force | Key Metric | Benchmark |
| Engagement | DAU/MAU ratio, D30 retention | DAU/MAU > 50%, D30 > 15-20% |
| Acquisition | Viral factor | 0.5 = 2x amp, 0.7 = 3.3x, 0.95 = 20x, >1.0 = explosive |
| Economic | CAC trend + LTV trend | Both improving simultaneously |
Escape Velocity Dashboard:
Engagement: DAU/MAU 35% | D30 28% | Trend up -> HEALTHY
Acquisition: Viral factor 0.6 | Amp 2.5x | Trend flat -> NEEDS WORK
Economic: CAC down, LTV up | Ratio improving -> HEALTHY
Weakest Force: ACQUISITION
-> "Engagement strong but users aren't inviting others.
At viral factor 0.95 you'd get 20x amplification vs 2.5x.
Focus on referral loop design."Identify weakest force and concentrate effort — applying TOC constraint logic to network growth.
Growth Ceiling Detection (5 Forces)
Monitor during periodic scans:
| Ceiling Force | Detection Signal | Response |
| Market saturation | >40% of addressable accounts in segment | "Adjacent user analysis needed. Who is the NEXT segment?" |
| Channel degradation | Response rate declining >30% over 3 months | "Channel degradation. Diversify or refresh approach." |
| Network revolt risk | Top 3 customers/creators = >50% of value | "High concentration risk. What if top creator leaves?" |
| Quality dilution | Engagement per user declining as base grows | "New users engaging 40% less. Onboarding or targeting issue." |
| Adjacent user barrier | Growth stalls despite marketing investment | "Current ICP saturated. Who is the adjacent user — aware but unable to engage?" |
Competitive Position Dynamics
| Share Split | Dynamic | Action |
| 50/50 | Stable coexistence | "No network advantage. Compete on product." |
| 55/45 | Slight advantage, reversible | "Small lead. Accelerate to 60+ before competitor responds." |
| 60/40 | Clear advantage, effects amplify gap | "Network effects in your favor. Invest in moat." |
| 70/30 | Overwhelming, reversal extremely difficult | "Dominant. Focus on preventing network revolt." |
T2D3 Benchmark
For SaaS with network effects, from $1M ARR: Triple -> Triple -> Double -> Double -> Double $1M -> $3M -> $9M -> $18M -> $36M -> $72M
Decision Rules
- Define atomic network first — no growth strategy without knowing the minimum viable network
- Density before breadth — anti-peanut-buttering rule is non-negotiable
- Zero rates are critical — track and minimize "zeroed" user experiences
- Weakest force gets focus — TOC constraint logic applied to the three forces
- Growth ceilings are early warnings — detect during growth, not after stall
- Share dynamics are exponential — small leads compound, act fast
Anti-Patterns to Detect
| Anti-Pattern | Signal | Response |
| Peanut-buttering | Spreading across 5+ segments simultaneously | "Achieve atomic network in ONE segment first." |
| Ignoring zeros | High zero rate with no action | "Users who get zeroed churn permanently. Reduce zero rate." |
| Force imbalance | Strong engagement, no viral loop | "Weakest force (acquisition) limits growth. Focus there." |
| Creator concentration | Top 3 = 50%+ of value | "Network revolt risk. Diversify value sources." |
| Premature expansion | Growing to new markets before first one works | "Atomic network not achieved in first segment. Don't expand yet." |
Other versions1 project uses a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 1
- deciqAI/knowledge-skills★ 11same name
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository24 other files it gives its agents.
- CLAUDE.mdCLAUDE.md
- anticipatory-sellingSkill
- choice-architectureSkill
- confidenceSkill
- content-rulesSkill
- critical-chainSkill
- culture-mapSkill
- disruptionSkill
- effectuationSkill
- failure-essenceSkill
- growth-metricsSkill
- kpi-treeSkill
- pipelineSkill
- plan-bSkill
- pre-mortemSkill
- qualificationSkill
- quest-interviewSkill
- retention-cohortsSkill
- review-engineSkill
- root-causeSkill
- strategy-kernelSkill
- stupid-gamesSkill
- unit-economicsSkill
- value-stickSkill
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
Reviews can't be read right now.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

