The Cold Start Problem by Andrew Chen

The Rabbit Hole is written by Blas Moros. To support, sign up for the newsletter, become a patron, and/or join The Latticework. Original Design by Thilo Konzok.

Key Takeaways

  1. There are five primary stages:
    1. The Cold Start Problem
      1. Who are the first, most important users to get onto a nascent network, and why? And how do you seed the initial network so that it grows in the way you want?
      2. When the Cold Start Problem is solved, a product is able to consistently create “Magic Moments.” Users open the product and find a network that is built out, meaning they can generally find whoever and whatever they’re looking for. The network effects kick in, and the market hits its Tipping Point as users start coming to you.
      3. How many users does your network need before the product experience becomes good? The way to answer this is for companies to do an analysis on the size of their networks (on the X-axis) plotted against a set of important engagement metrics (on the Y-axis).
      4. For new products, it’s important to have a hypothesis for the size of your network even before you begin. Communication apps can be 1:1, so the network is small, and you can plan accordingly. Contrast that to products that are highly asymmetrical, with content creators and viewers, or marketplaces with buyers and sellers—these are likely to require a much bigger number to hit the threshold, and require a much bigger effort to get started. The size of an initial network helps determine a launch strategy.
      5. Density and interconnectedness is key.
      6. The solution to the Cold Start Problem starts by understanding how to add a small group of the right people, at the same time, using the product in the right way. Getting this initial network off the ground is the key, and the key is the “atomic network”—the smallest, stable network from which all other networks can be built.
      7. My advice: Your product’s first atomic network is probably smaller and more specific than you think. Not a massive segment of users, or a particular customer segment, or a city, but instead something tiny, maybe on the order of hundreds of people, at a specific moment in time.
      8. First, the concept of atomic networks provides the clearest goalposts for an upstart network—it’s all about splitting off, or creating from scratch, a distinct and higher-density atomic network. Network density beats total size, a theme we’ve seen throughout the examples of this book.
    2. Tipping Point
      1. Generally one side of the network will be easier to attract—this is the easy side of the network. However, the most important part of any early network is attracting and retaining “The Hard Side” of a network—the small percentage of people that typically end up doing most of the work within the community.
      2. Not only does a product have to appeal to the hard side of the network, but as I discuss in “The Killer Product,” the most successful network effects–driven apps are also sometimes dead simple. They eschew a long list of features and instead emphasize the interactions among people using the app.
      3. Thus the order of operations, at least for most consumer-facing marketplaces, is “supply, demand, supply, supply, supply.” While supply might be easy to get onto the network early on through subsidies, eventually it will become the bottleneck.
      4. The key insight in the stories of Homobiles or Tinder is—how do you find a problem where the hard side of a network is engaged, but their needs are unaddressed? The answer is to look at hobbies and side hustles.
      5. The Magic Moment is a nice concept, but it would be even more useful if you could measure it. The way to best do this might be surprising—you start with the opposite of magic, the moments where the network has broken down, and you start solving the problem from there. At Uber, we called these moments Zeroes. A zero at Uber was the worst experience you could have, when a rider opens the Uber app with the intent to pick an address and pick up a ride—but there aren’t any drivers in the area! This is a zero. When the point of the product is to interact with other participants in the network, a zero means that its value can’t be fulfilled, which means users will bounce and possibly never come back again.
      6. For networked products, the curation of the network—who’s on it, why they’re there, and how they interact with each other—is as important as its product design. Starting with a deliberate point of view on who’s best for your network will define its magnetism, culture, and ultimate trajectory.
      7. People You May Know was a key part of LinkedIn’s success, generating billions of connections within the network. It started with “completing the triangle”—if a bunch of your friends have all connected with Alice but you haven’t yet, then there’s a good chance you might know Alice, too.  This helped scale the density of the LinkedIn network so that even after you added hundreds of connections, the site could still help recommend relevant people to you. This is a direct example of alleviating the overcrowding dynamics of a social network, which is exactly why people recommendations, relevance-driven feeds, trending topics, and a slew of other algorithmic approaches have been layered onto social products over time.
    3. Escape Velocity
      1. The Escape Velocity stage is all about working furiously to strengthen network effects and to sustain growth. This is where the classical definition of a “network effect” is wrong. I redefine it so that it’s not one singular effect, but rather, three distinct, underlying forces: the Acquisition Effect, which lets products tap into the network to drive low-cost, highly efficient user acquisition via viral growth; the Engagement Effect, which increases interaction between users as networks fill in; and finally, the Economic Effect, which improves monetization levels and conversion rates as the network grows.
      2. The Engagement Effect manifests itself by increased engagement as the network grows—this can be developed further by conceptually moving users up the “engagement ladder.” This is done by introducing people to new use cases via incentives, marketing/communications, and new product features. Uber did this via leveling users up from airport trips to dining out to daily commutes.
        1. Content and communications might be a series of how-to videos teaching effective use of LinkedIn’s connection features. And an incentive might look like a free subscription when the user completes certain actions. A product road map can be generated with hundreds of these ideas, large and small, and then prioritized.
        2. Answering these questions generates a long stream of potential experiments and ideas to try. It’s incredibly useful to lay out an engagement loop, one screen at a time, and brainstorm ways to increase each step—this method is at the heart of what I typically do when advising startups on creating higher stickiness.
      3. “The Acquisition Effect,” on the other hand, is the network effect that powers the acquisition of new customers into your product—in other words, viral growth. Products are inherently viral when people bring their friends and colleagues into a network simply by using it—as Dropbox, messaging apps, and social networks do.
        1. As a rough benchmark for evaluating startups at Andreessen Horowitz, I often look for a minimum baseline of 60 percent retention after day 1, 30 percent after day 7, and 15 percent at day 30, where the curve eventually levels out. It’s usually only the networked products that can exceed these numbers. That’s because networked products are unique in that they often become stickier over time, which cancels out the inevitable customer churn. In rare but exceptional cases, the curve will “smile”—meaning that retention and engagement will actually go up over time, and churned users will reactivate. I’ve learned that when a startup shows a smile curve, you should probably try to invest. It’s exceedingly rare.
      4. Economic Effect means that the leading network often has a better business model. Products with a strong Economic Effect are able to maintain premium pricing as their networks grow, because switching costs become higher for participants who might be looking to join other networks.
    4. Hitting the Ceiling
      1. Rather than focusing on the core network of Power Users—the loud and vocal minority that often drive product decisions—instead the approach was to constantly figure out the adjacent set of users whose experience was subpar.
      2. In the framework of adding layers to a cake, serving each adjacent network is like adding a new layer. Doing this requires a team to think about new markets, rather than listening to their vocal core markets—a hard feat when the core market generates most of the revenue. For the core market, there’s a different way to grow: adding new formats for people to connect and engage with each other.
    5. The Moat
      1. In the early days at Airbnb, we would always talk about creating a positive “Expectations Gap.” In the early days, when we were new, guests go in with low expectations, but then would be blown away by the experience. You need this high NPS to get people to tell their friends, and it makes hosts more likely to join too. Our competitors who took shortcuts couldn’t deliver here.
      2. Rarely in network-effects-driven categories does a product win based on features—instead, it’s a combination of harnessing network effects and building a product experience that reinforces those advantages.
      3. Focusing on the hard sides of the network, which are usually smaller in number, provides leverage in competitive moves.

What I got out of it

  1. A pragmatic deep dive on what the cold start problem is, how to conceptualize it, how to try to overcome it, and if you’re lucky, how to ride it