Ask the Wizard

When asked what kinds of consumer internet companies investors like to fund, the most common refrain from the investors themselves is "companies with network effects". For those of you who still haven't unwrapped your consumer internet home game, that great seer of all things, the wikipedia, has a nice entry on the subject.

Investors like to see network effects, in theory, because the implication then is that even if potential competitor XYZ builds a better mousetrap, their lack of customers makes it difficult at best for them to gain a beachhead against you. eBay vs. Amazon and Yahoo auctions was one of the best examples of this in action. Despite on par features and huge resources they could bring to bear, Yahoo and Amazon were unable to provide sellers with the marketplace size of eBay. The sellers go where the buyers go, which attracts more of the buyers to where everything's being sold, and better features and functions don't matter against a more liquid marketplace. One could make the same argument about AdSense these days. There are ways to compete against network effects, but one of those ways generally isn't "charge the front gate of the castle".

So, when you're thinking about your company's good or service, it would be helpful to think about how you could achieve network effects, right? Eh, I don't think so. I think network effects accrue to most companies by accident, as an artifact of something they were doing that led to network effects. It's not obvious in the early days of the company where any potential network effects will come from, and even if it IS obvious, it's not at all clear that you can leverage these to your advantage. A good example of this last point is an Instant Messaging platform. Huge network effects - if I'm using AIM and Fluffy is using AIM and we both know Wiley, then it doesn't make lots of sense for Wiley to use Jabber. At least, that's the pitch. In reality what happened is that wiley used AIM and Jabber and Yahoo Messenger and then Fluffy and I started using them all as well. The first mover in a new communications space can generally monetize network effects by selling quickly before they have to drive profits (see Mirabolis/ICQ in instant messaging and Skype in VOIP), but it's not at all clear that the first mover has any long term strategic advantage that could drive value (aka profits).

So, if it doesn't make sense to spend lots of time thinking about the strategic advantage you can gain from network effects, what should you think about when you're trying to understand the potential competitive landscape for your company? Stay tuned for Strategic Advantage, Part II, in which I will describe my favorite way of thinking about strategic advantage: Quantum Hidden Barriers to Entry.