This is a long and market-y way to basically explain two ideas. Here's the simple idea: Viruses infect new users. Users go up. Going viral would be things lik

This is a long and market-y way to basically explain two ideas. Here's the simple idea: Viruses infect new users. Users go up. Going viral would be things like FOMO or hijacking a popular twitter account with a sign-up bonus link. A network effect is sometimes used to describe interaction between users. Users go up because users stay or users don't leave. There's a stark difference in an app that everybody uses versus an app that nobody uses, that's basically it. Math wise: A virus forms a graph that is a tree. Nodes infect nodes infect nodes. Nodes have to be uninfected to be infected. Lots of real life examples. A network effect is a network of nodes that interact with each other. This is often just like the example they give, where nodes can form multiple connections to each other. If each edge connecting a vertex represented a function adding value to each user, and a viral effect can only sustain an edge to a vertex it previously was not connected to, then yeah by definition one has more connections than the others. All this article is trying to say, a viral effect can only add value from one user to another. A network effect can add value from one user to many users and likewise, many users can add value to one user. One-to-one vs Many-to-Many