Joachim Andersen

Accelerated Learning

Network Effects: Why Value Grows With Every New User

About fifteen antique brass candlestick telephones stand on a polished black marble table at night, their golden cords linked from each telephone to every other into a glowing web
Every new connection makes the whole network more valuable.


Network Effects describe how a product, service or idea becomes more valuable to each user as more people use it. A single telephone is useless, but every new person who joins a network makes it more useful for everyone already in it, which in turn attracts even more people.

The essence of Network Effects is to ask: “Does this become more valuable the more people use it, and who is it pulling in next?”


Origin: One System, Universal Service

In the early 1900s, American telephones were split between Bell and many independent companies, whose networks were often not connected to each other. In the 1908 annual report of AT&T, its president Theodore Vail presented the idea that a telephone’s value grows with the number of people it can reach. Under the slogan “One Policy, One System, Universal Service”, most of the more than 4,000 local and regional exchanges of the time were eventually merged into a single Bell System.

Economists began studying the idea in earnest in 1974, when Jeffrey Rohlfs at Bell Labs showed that the demand for a communications service depends on how many others already use it, which means a new service can fail simply because too few people join at the start. In 1985, Michael Katz and Carl Shapiro called these benefits network externalities, and the term network effect soon became central to how we think about technology markets.


Key Concepts

  • Direct and Indirect Effects
    A direct network effect happens when more users make the product better for other users, like a messaging app. An indirect effect happens when more users attract more complementary products, such as apps for a phone or movies for a video format, which then attract even more users.

  • Two-Sided Markets
    Many networks connect two different groups, like buyers and sellers or drivers and riders. Each side only wants to join if the other side is already there, which creates a chicken-and-egg problem that has to be solved before the network can grow.

  • Critical Mass and Tipping
    Below a certain size, a network is worth too little for people to join, so it can stall or shrink. Above it, growth feeds itself, and the market often tips toward a single winner, even if a competing product is just as good or better.

  • How Fast Value Grows
    In the early 1980s, Robert Metcalfe suggested that the value of a network grows with the square of the number of users, which became known as Metcalfe’s law. In 2006, Bob Briscoe, Andrew Odlyzko and Benjamin Tilly argued that this overstates it, since most possible connections are worth little, and that value grows more slowly, but still faster than the number of users.


Some Examples

  • The Fresno Drop
    In September 1958, Bank of America launched its BankAmericard in Fresno, California, by mailing 65,000 unsolicited credit cards to its residents. With people already holding cards, shops had a reason to accept them, and with shops accepting them, people had a reason to use them. Within about a year, more than 2 million cards had been sent out across California and 20,000 merchants accepted them. The launch was plagued by fraud and losses, but the network it created later grew into Visa.

  • The Videotape Format War
    Sony launched Betamax in 1975, a year before JVC’s VHS, and many considered it the better format. But VHS offered longer recording times and was licensed to many manufacturers, so more households bought it. As its market share grew, movie studios and rental stores stocked more VHS tapes, which made VHS players even more attractive. Sony started selling VHS recorders itself in 1988 and made its last Betamax machine in 2002.

  • The Fax Machines That Wouldn’t Die
    Early fax machines from different makers often couldn’t talk to each other. In 1980, a shared international standard let any fax machine send a page to any other in under a minute, which led to two decades in which almost every office needed one. Because everyone else had one, they stayed long after email arrived, and in Japan they were still widely used in 2025, even in convenience stores.

  • The Unraveling of Friendster
    Launched in 2003, Friendster reached 3 million users within months and turned down a $30 million offer from Google. As Facebook and other rivals grew, users in the United States began to leave, and each one who left made the site less valuable for their friends. Its traffic there fell sharply from 2009, and it was sold that year for $26.4 million. In 2013, David Garcia, Pavlin Mavrodiev and Frank Schweitzer studied its collapse and showed how such departures can cascade through a network once the cost of staying outweighs the benefit of the remaining friends.


Applying Network Effects

When I want to use this mental model, this is the simple process I follow:

  1. Spot the Network
    When I evaluate a product, a company, a platform or even a community, I ask whether it becomes more valuable as more people join, and whether that value comes from direct or indirect effects.

  2. Look for Critical Mass
    When I start something new that depends on others, I focus on getting a small group of people to find it valuable first, rather than trying to reach everyone at once.

  3. Solve One Side First
    If it connects two groups, I ask which side is hardest to attract and how I can make it worthwhile for them even before the other side arrives.

  4. Choose Standards Carefully
    When I pick a tool, a format or a platform, I consider what others around me use, because being compatible is often worth more than being slightly better.

  5. Watch for the Reverse
    When a network I depend on starts losing people, I take it seriously early, because decline can speed up just as quickly as growth did.

  6. Build My Own Network
    I treat my relationships and my audience as networks too, where each new connection makes the existing ones more valuable, and where giving people reasons to connect with each other helps everyone.


Pitfalls and Limitations

  • Not everything has network effects. Many products don’t get better as more people use them, and calling something a network effect doesn’t make it one.

  • Bigger isn’t always better. Crowding, spam and noise can make a network worse as it grows, and many connections are worth little or nothing to me.

  • The best product doesn’t always win. Networks can lock us into a worse standard, simply because switching would mean leaving everyone else behind.

  • Networks can collapse. The same feedback loop that made a network grow can make it shrink, and a network that seems unbeatable can unravel surprisingly fast.


Related Mental Models

Network Effects remind me that the value of many things depends less on what they are and more on who else is using them. By spotting where value grows with each new user, paying attention to critical mass and watching for signs of decline, I can make better choices about what to build, join and rely on.



To learn more about mental models and philosophy,
check out: Wise: How to Attain a Deep Understanding of the World, History, Our Place in It, and How to Live Well



📚 Sources and References