(2019-07-29) Hidden Networks Network Effects That Dont Look Like Network Effects

Hidden Networks: Network Effects That Don’t Look Like Network Effects. Here are three(-ish) examples of networks that don’t look like networks.

1) Slow networks: Slow networks are defined by a time delay between when the network is created and when it starts to show value.

Take Lambda School

The advantage of slow networks is that once built, they’re usually very difficult to displace. The top universities took hundreds of years to build their network effects, and despite periodic predictions of their demise, these networks effects don’t seem to be showing any sign of weakening.

Another example of a slow network is social lending

When my co-founders and I started Frank

borrowing money happens relatively infrequently (about once every three years for our users)

other categories like hiring, healthcare, or real estate

2A) Unfinished networks: An unfinished network is one where a product feature or strategic decision leaves the network temporarily incomplete in some material way. When the network is eventually completed though, the network effects are immediately clear.

One successful example of an unfinshed network from a few years ago was OpenTable. In its earliest days, OpenTable looked more like a SaaS business than a network effects one.

OpenTable spent its first five years of existence primarily focused on signing up restaurants.

The challenge, of course, is that unfinished networks often don’t get completed

2B) Throttled networks: A throttled network is one where a product feature or strategic decision limits the size or engagement of the network in a material way, thereby masking the strength of the network effects.

Often times these communities are actually an audience rather than a latent network, meaning users derive value from the central node, not the network. When it’s just an audience, the tool or product ends up scaling more like a linear business (e.g., a DTC product) than a network

Look to see if the network’s users are engaged with each other or just the central node

3) Latent networks: (aka “Come for the network, stay for the tool” networks)

The concept is that you start by building a community that acts like a network, with users interacting, engaging, and generally creating value for each other. Eventually, a product is introduced that catalyzes or amplifies the way the network engages.

Savvy game developers have been following this playbook for years. Even before there is a game to test, they’ll set up a Discord server to help build the community and network of gamers. Most importantly, this ensures there is a vibrant ecosystem when the game launches, which is important in social games where having other players materially improves the experience. Hypixel and Phoenix Point are a couple of examples where they started by building robust communities that eventually became (or will become) the networks that are layered on top of the games.

One example is executive networks. Chief is an executive network for women. Think of it as Young Professionals Organization (YPO), but focused on women in the C-suite.

At the end of the day, building a network effects business is essentially a race. One company is trying to get to that magic tipping point before their competitors.

in the long run, they might unlock the value of their network through increasing opportunities to engage, pricing that rises with network value, or even a deeper membership pool

It’s an important distinction to separate throttled networks and exclusive ones. Throttled networks are, in a certain sense, temporarily small. The value proposition can support a larger network, just not yet. In contrast, networks that rely on exclusivity — dating apps like Raya or members clubs like Soho House are a couple of examples here — often have an upper limit to their network effects.

Hidden networks are hidden advantages.