The Cold Start Problem: How to Start and Scale Network Effects

by Andrew Chen

When one network wins at the expense of the other, the reemergence of the Cold Start Problem makes it difficult to ever recover that network.

Platform dependence can be disastrous if not managed well. If you integrate too closely with a preexisting network, allowing them to control your distribution, engagement, and business model, you become just a feature of their network.

The largest networks can take a lot, in many networks, but they remain vulnerable to any new upstart that uses cherry picking as a core strategy.

While this might superficially look like a large user base, it actually consisted of many weak networks that weren’t engaged, because most new users showed up and tried out the product as they read about it in the press, rather than hearing from their friends. The high churn in the product was covered up by the incredible fire hose of traffic that the rest of Google’s network generated.

When unengaged users interact with a networked product that hasn’t yet gelled into a stable, atomic network, then they don’t end up pulling other users into the product.

When networks are built bottom-up, they are more likely to be densely interconnected, and thus healthier and more engaged.

here’s the paradox: To build a massive successful network effect, I argue that you must start with a smaller, atomic network. And use the success in the first set of networks to tip over the next set of small networks.

Focusing on the hard sides of the network, which are usually smaller in number, provides leverage in competitive moves.

the product really matters, and bundling can provide a huge distribution advantage, but it can only go so far.

consider each of the Engagement, Acquisition, and Economic network effects—while new user acquisition can be propped up by cross-selling users from one product to the other, the engagement and monetization effects will only kick in when there’s a real critical mass.

The move is to leverage a larger network across multiple touchpoints, to accelerate all of its network effects—not just acquisition.