Network effects

Not every "network" is a moat. NFX's Network Effects Bible classifies them so you can diagnose whether you have one or are just hoping for one. A SaaS tool that "shares dashboards" is usually just collaboration. If a user gets nearly full value alone, you do not have a network moat — you have a product.

The types

• Direct (same-side): each additional user makes the product more valuable to users of the same type (messaging, social graphs).
• Indirect (cross-side): more of group A attracts group B, which attracts A (marketplaces, platforms, app stores).
• Data network effects: more usage produces better models or rankings that improve the product for everyone (search, recommendations, some fraud systems).
• Local / personal: value is concentrated in a subgraph (your coworkers, your city, your friends) — often easier to ignite than a global network.
• Marketplace liquidity: density of matched supply and demand in a category or geography; below a threshold the market feels empty.

Walking through local density

A two-sided marketplace launches city-by-city. In City A it reaches 500 riders and 80 drivers; median wait is 4 minutes and drivers earn enough to stay online. In City B it has 80 riders and 20 drivers; waits are 18 minutes and drivers churn.

Same product, different liquidity. The moat is local density, not the app binary. Expanding to City C before A and B are dense burns cash without creating a network that defends.

False network effects are common. Ask: if this user were the only user, how much value remains? If the answer is "almost all of it," you are selling software, not a network.

Data network effects need a similar test: does more usage produce data a rival cannot buy at similar cost? Public-web training data usually fails. Labeled outcomes from your workflow — which tickets resolved, which recommendations converted — can pass.

Common mistakes

• Calling collaboration a network effect. Shared dashboards are a feature.
• Expanding geographies before local liquidity. You burn cash and still look empty.
• Claiming a data network on public data anyone can scrape.

Check yourself

City A: 500 riders, 80 drivers, 4-minute waits. City B: 80 riders, 20 drivers, 18-minute waits. Where is the moat?

City A's local density. The binary is the same; liquidity is not.

A project-management tool lets you invite teammates. Network effect?

Usually no. One user still gets most of the value. That is collaboration, not a network economy.

Key takeaways

• Diagnose the type: same-side, cross-side, data, local, liquidity.
• If a user gets full value alone, you do not have a network moat.
• Local density is often the real marketplace defense.
• Data effects require data a rival cannot buy at similar cost.

Further learning

The Network Effects Bible (NFX)Network effects have been responsible for 70% of all the value created in technology since 1994. Founders who deeply understand how they work will be better positioned to build category-defining companies.NFX

Next, aggregation theory.