Many founders claim their startups contain network effects, but many of these claims only scratch the surface. Networks that achieve greatness are designed, with many trade-offs and factors in mind regarding the "nodes," how interactions take place, what causes people to stay engaged, and the types of relationships that get created. Anu Hariharan shows us the specific decisions and trade-offs we can weigh that startups often overlook.
Anu Hariharan:
If you're a founder in year one or year two, your job is to first get 100 users that love you, 1,000 users that love you. But by year two, the question to ask is what is the driver of critical mass? And then using that as an algorithm to build scale.
Mike Maples:
This comment by Anu Hariharan is totally on point when it comes to a key lesson of greatness. Great networks are designed intentionally. Let's talk about how.
Mike Maples:
Network effects have become the most important way to build a competitive mode in today's world. What's the big deal about network effects and what can you do to increase the chance that you can make them central to your startup's path to greatness? Let's start from the beginning. The first thing to understand about network effects is that they're extraordinarily powerful. From classic economics, network effects create a profound benefit, both to customers as well as a formidable barrier from competitors for the network operators who own the networks in the first place both at the same time. The benefit is that each user on the network is getting more value from the fact that all the other users are on the network at the same time, which means that the leader can theoretically increase their prices as they gain scale, which was previously unheard of in business. The barrier is that a competitor must spend a potentially insurmountable amount of money to catch up to the leader.
Suppose I decided to create a new social network that competes directly with Facebook and I want all of its members to jump ship from Facebook and join my network instead. Well, because of people's already established connections, free wouldn't be good enough as a value proposition. I would literally have to pay people by the billions all at the same time to leave. So as a result, the best network effects businesses frequently demonstrate winner-take-all characteristics in their chosen markets, and this is what you want to see as a founder pursuing greatness. The second thing to understand about networks is they don't just happen, they're designed. This is where your decision to be great matters most. Being thoughtful and strategic about the network you're building is essential from day one and that is the focus of this lesson of greatness. When you design your network, there are three key elements you must consider.
The first element is how you get your network started and succeed at getting it to spread. There are several key things to get right in kick-starting and accelerating your network. First, you want to know what is the primary value proposition for your users. Can value be created in single player mode at first, like with Facebook and its initial efforts to create a complete student directory? Or is the product useless without operating in multiplayer mode like Slack or Skype? Second, you want to consider the most effective growth tactics. How will you get your network to spread organically without spending a fortune on marketing programs? Third, you should try to isolate your engagement trigger. What is the threshold that a customer should cross to likely become a fully engaged customer? And finally, you should identify the primary thing that will further accelerate your network effects. What will make the flywheel accelerate?
Perhaps it's easiest to show with a few examples. Let's start with Facebook, which is a classic network. Facebook started with a single player mode value proposition. It was an online student directory that was valuable for each individual student. Their growth tactic was exclusivity and word of mouth. Initial access was given to the entire Harvard student population, which was critical for early adoption. And because the product was inherently viral, it spread from one user to the next as an inherent result of people using it in the first place. Friends wanted to friend other friends and so it spread. In terms of their engagement trigger, Facebook figured out that connecting a new user to 10 new users within 14 days of signup was critical to keeping them engaged. So they used tactics like email contact import, suggested friends, and embedded widgets to drive that engagement particularly in the early days of use. They built accelerating network effects by tweaking their product to increase engagement for each person on the network.
Examples were relationship status, timelines, and other features that ensured that individuals would use the product more even as the size of the network grew. Now let's consider Airbnb marketplace. Airbnb also had a single player mode value proposition. For guests, they offered a place to stay for 30% to 80% less money when hotels in a given city were completely booked. For hosts, they offered a way to rent their property to someone in the height of the great recession. Their growth tactic was to target cities with sold out events and constrained hotel supply. For example, they made a big push to build awareness in Denver during the Democratic Convention in 2008 when Obama was the nominee and the hotels were massively sold out. In order to gain critical mass, they launched photography services to make host offerings more appealing. They also offered information about mutual connections and other guest reviews to further enhance trust. In order to accelerate network effects, they focused on increasing the number of guest room nights.
The more hosts who got guests, the more people wanted to be hosts. And the more hosts who were available, the more appealing it became for people to want to become guests. Notice that in these examples, we described different of networks, which is the second key element to successful network design. You want to think about the type of network you want to build? Is it a classic network like Facebook, IE a social network, telephone network, or office printer network where people come together to interact or share things? Or is it a marketplace like Airbnb that creates a way for buyers and sellers to come together like online auctions or even dating sites? Or is it a platform like an operating system or a messaging app, like WeChat? Ultimately your choice of what type of network you wish to build is key to getting the correct design.
If you're building a classic network, you want to know things like what should the entry point be to build a network effect? What are the growth levers, tactics, and hacks to get to critical mass? What is the critical mass inflection point at which a network effect occurs? How do you drive engagement? How do you take advantage of irregular typologies to find clusters and sub-clusters? For a marketplace, you have additional considerations. How do you build liquidity in the marketplace? How do you solve the chicken and egg problem of getting enough supply and demand at the same time? Which is the money side versus the subsidy side of the marketplace? For a platform, you need to ask additional questions.
Will the market we care about be served by a single platform or multiple platforms? Will the platform that wins be shared like the ethernet or will they be a fight for proprietary platform control like Microsoft versus Apple? And Anu shows us that we can go even further in our thoughts about network design. We can drill down even further and get even more precise about nodes and clusters, which describe the interconnections of people in the network. This is where network properties come in. Should the individual nodes in your network be homogeneous and all play the same role like peer-to-peer calls on Skype or should they be more heterogeneous with different roles like Airbnb with hosts and guests?
What type of clustering do you think will work best, hub and spoke, small connected groups? What directionality will you have with the connections? Should you be unidirectional like Twitter with its follower model or does it make sense to be more like Facebook where friends have bi-directional connections? What about potential compliments to your network? A complimentary product increases the usage of your product when it's used if you tie them together properly. A great example of this is the relationship between Microsoft Windows and Microsoft Office. Operating systems like windows have a strong network effect of their own, because they attract a critical mass of developers and users. And the productivity apps in Office also attract users to the Windows platform creating a virtuous cycle.
The same is true with the App Store and the iPhone. And of course, when you start to design and operate your network, there are important things to watch out for. The biggest mistake is confusing viral growth with network effects. They are not the same thing. Network effects are about creating value for users as more users join the network. But virality is about the speed of adoption of a network service. It's more about distribution than about product. They might happen at the same time, but they should not be conflated. This is important because not all network services are viral right away even the legendary ones. Facebook had network effects and virality. Obviously we know how that turned out. But Airbnb, it turns out had network effects, but lacked virality and did just fine. They just had to build the network differently. So that's the key thing. Both networks are great, but they had to be designed differently given their characteristics.
So what does this mean for you? Network effects are the most powerful economic force of our time. When a startup executes a network effects strategy well, they create a value proposition that can't be matched from the customer perspective or defeated from the competitive perspective. If you're starting a startup, you should always try to find ways to design valid network effects in your products. Don't treat it as an afterthought. And you should also be precise in your thinking and communication to employees, partners, investors, and everyone involved about your design decisions. Getting your thought process right and acting on it is a fundamental part of your opportunity to find a path to greatness.
Thanks for listening to this episode of Greatness. If you found this episode insightful, you might also enjoy my interview with Reed Hoffman, who I like to call the Network Philosopher King. In it, we talk about how Reed built LinkedIn is a network from the ground up by design. Here's a preview of Reed's insights.
Reed Hoffman:
What most matters to other people? People. Sure, they like places. Sure, they like things. But we're social animals, focus on people. How do you make people better with each other? That's the thing to do and that's part of where LinkedIn came from.
Mike Maples:
If you're new to the science of network effects, it might seem like I just threw a lot at you in this episode. But the good news is that you can develop a deeper understanding of these ideas very quickly and they'll improve your chances of success dramatically. In order to help those who want to go in more depth, I put a bunch of links and other content in our newsletter about network effects. You can check them out by going to greatness.substack.com. I appreciate you listening and I'd love to have you subscribe wherever you get your podcasts so you don't miss an episode. And if you like the show, I'd be grateful if you could leave us a review on Apple Podcasts. And until we catch up again, I hope you never let go of your inner power to do great things in whatever matters to you. Thank you for listening.