The Cold Start Problem
by Andrew Chen (2021)
Key Takeaways
- ✓ The cold start problem is the chicken-and-egg challenge every network faces -- the product has no value without users, but users will not join without value, and solving this paradox is the first and most critical challenge
- ✓ The atomic network is the smallest possible network that can deliver value -- finding and building your atomic network is more important than pursuing broad growth because a small, dense network creates more value than a large, sparse one
- ✓ The tipping point occurs when a network becomes self-sustaining -- growth becomes organic rather than subsidized, and the network effects begin compounding, but this inflection point requires reaching a critical density threshold
- ✓ Network effects can work in reverse -- the same dynamics that drive growth can drive decline when key users leave, content quality drops, or the network becomes overcrowded, and managing these forces is as important as building initial growth
- ✓ Hard and easy sides of a network require different acquisition strategies -- the hard side (drivers for Uber, creators for YouTube, sellers for marketplaces) constrains growth and must be cultivated with different incentives than the easy side
4/5
Andrew Chen, general partner at Andreessen Horowitz, provides the definitive guide to building and scaling network effects. The book addresses the fundamental chicken-and-egg problem that plagues every networked product -- how do you attract users when the product is only valuable once other users are already there? Drawing on examples from Uber to Slack to Tinder, Chen maps the complete lifecycle of network effects...
The Definitive Guide to Network Effects
The Cold Start Problem fills a surprising gap in business literature. Network effects are arguably the most powerful force in technology business models, but until this book, there was no comprehensive, practical guide to building them. Chen draws on his experience as a growth advisor (he wrote the viral essay on the growth of Uber) and his role at Andreessen Horowitz to map the complete lifecycle of network effects from cold start to maturity to potential decline.
The book is organized around five stages: the cold start problem, the tipping point, escape velocity, the ceiling, and the moat. This lifecycle framework gives founders and product leaders a roadmap for understanding which stage they are in and what challenges to expect next.
Solving the Cold Start
The cold start problem is the fundamental paradox of network products: the product has no value without users, but users will not adopt a product that has no value. Every network product — from Airbnb to Slack to dating apps — must solve this paradox before anything else matters.
Chen’s key insight is the concept of the atomic network — the smallest possible unit of the network that can deliver standalone value. For a messaging app, this might be a single friend group. For a marketplace, it might be one city or one product category. For a workplace tool, it might be one team within one company.
The strategic implication is to stop trying to build a large network from scratch. Instead, identify your atomic network and build it one unit at a time. Uber launched city by city, not nationwide. Slack spread team by team within organizations. Tinder started at one college campus. Each atomic network provides enough value to retain users, and the collection of atomic networks eventually forms a larger network.
The Hard Side Is the Constraint
Chen introduces the concept of hard and easy sides of a network. In a marketplace, the hard side is the supply side (the side that requires more effort, investment, or commitment to participate). For Uber, drivers are the hard side. For YouTube, creators are the hard side. For Airbnb, hosts are the hard side.
The hard side constrains growth because it is harder to recruit and retain. Most failed marketplace startups fail because they cannot solve the hard side. They attract demand (easy) but cannot attract or retain supply (hard). The strategic implication is that hard-side acquisition should receive disproportionate attention and investment.
The strategies for recruiting the hard side are different from those for the easy side. The hard side often needs economic incentives (Uber guaranteed earnings for early drivers), tools and support (YouTube provided analytics and monetization), or status and community (Instagram featured early photographers). Understanding what motivates the hard side of your specific network is essential.
The Tipping Point and Escape Velocity
The tipping point occurs when a network becomes self-sustaining — organic growth exceeds subsidized growth, and the network effects begin compounding. Before this point, growth requires constant investment. After it, growth becomes increasingly efficient.
Reaching the tipping point requires achieving critical density in your atomic networks. A dating app with fifty users in a city is useless. The same app with five thousand users in a city is viable. The threshold varies by network type, but the principle is consistent: density matters more than breadth.
Escape velocity is the stage where multiple growth mechanisms compound. Viral growth, paid acquisition, content marketing, and network effects all reinforce each other. Companies at this stage grow faster than at any previous stage, and the growth itself strengthens the network effects that drive future growth.
The Ceiling and Decline
Chen’s most differentiated contribution is his treatment of network effects in reverse. The same dynamics that drive growth can drive decline. When key users leave, the network loses value for remaining users, which causes more departures. When content quality degrades (as networks attract lower-quality participants at scale), engaged users leave first, further degrading quality.
Managing these negative network effects is as important as building positive ones. Chen discusses strategies for maintaining quality as networks scale, managing the tension between growth and curation, and recognizing the warning signs of decline before they become irreversible.
Read This If…
You are building a product with network effects — a marketplace, social network, communication platform, or any multi-sided product — and need a practical framework for each growth stage.
Skip This If…
Your product does not rely on network effects. The book’s frameworks are specific to networked products and do not translate well to traditional software, services, or physical products.
Start Here
Read the cold start chapter and the atomic network concept first. These solve the most immediate problem for early-stage network products. Then read the hard side chapter, which addresses the most common failure mode. The ceiling chapters are worth reading if your network is already at scale.
Get This Book
Links may earn us a commission at no extra cost to you.
Related Reading
Zero to One
Peter Thiel, co-founder of PayPal and Palantir, argues that the most valuable businesses create something entirely new rather than copying what already exists. Drawing on his experience as an entrepreneur and investor, Thiel challenges founders to think about building monopolies, embracing contrarian thinking, and creating the future...
The Lean Startup
Eric Ries introduces a scientific approach to creating and managing startups, arguing that the biggest risk for new ventures is not building the wrong product but building something nobody wants. Through the Build-Measure-Learn feedback loop and the minimum viable product, Ries offers a systematic methodology for navigating uncertainty...
The Tipping Point
Malcolm Gladwell examines how ideas, products, and behaviors spread through populations like epidemics. By identifying the three rules of epidemics -- the Law of the Few, the Stickiness Factor, and the Power of Context -- Gladwell explains why some trends take off while others die...
The Hard Thing About Hard Things
Ben Horowitz, co-founder of Andreessen Horowitz, shares brutally honest advice about the hardest problems in running a startup -- the problems that business schools do not teach. From firing friends to managing your own psychology, Horowitz draws on his experience as a wartime CEO to offer practical wisdom for the worst days of entrepreneurship...
Good to Great
Jim Collins and his research team spent five years analyzing companies that made the leap from good to great, identifying the key factors that distinguish truly exceptional organizations from merely competent ones. The findings challenge conventional wisdom about leadership, strategy, and corporate culture...
Enjoyed this insight?
Get weekly book insights and reading recommendations.