How Network Effects Actually Work (and When They Don't)
Network effects are the most invoked concept in technology investment, and the most frequently misunderstood. Every pitch deck claims them. Very few businesses actually have them. Understanding the difference between genuine network effects and network effect theater is one of the most useful analytical skills in technology and business.
The actual definition
A network effect exists when a product or service becomes more valuable to each user as more users join. The telephone is the canonical example: the first telephone was useless, the second made two people able to communicate, and each additional telephone increased the value for all existing users.
Note what this definition requires: not just that scale provides benefits, but that each additional user makes the product more valuable for existing users. Many businesses that claim network effects actually have economies of scale (lower unit costs with more volume) or data advantages (better recommendations with more data). These are real competitive advantages. They're not network effects. Conflating them leads to significant analytical errors.
Types that actually hold
Direct network effects (more users directly benefit existing users) are the strongest and rarest. Social networks, communication platforms, and marketplaces with direct buyer-seller matching have genuine direct network effects. LinkedIn is more valuable with 900 million members than with 100 million because the person you need to reach is probably on it.
Indirect network effects (more users of one side benefits the other side) are more common but more fragile. App stores benefit from more apps, which attracts more users, which attracts more developers. These two-sided dynamics can be strong, but they're also vulnerable to the side with the lower switching costs defecting when a better option appears.
When network effects fail to protect
Network effects are a moat, not a wall. They slow competitive entry but don't stop it. MySpace had network effects. So did BlackBerry's BBM. So did Internet Explorer. Each was eventually displaced by a product that was sufficiently better to overcome the switching cost.
The historical pattern is: dominant platforms are most vulnerable when a new technology creates a new usage context rather than competing directly. Facebook didn't displace MySpace by being a slightly better social network. It offered a different product (real identity, college social graph) in a new context (broadband dorms). Competitors that try to replicate what you have lose to the network effect. Competitors that make what you have irrelevant don't have to fight the network at all.