Competition Matters Less in Software

What is it that makes software startups so able to disrupt incumbents? Why does competition seemingly matter less in software compared to other industries?

Software companies are very maneuverable. Especially new entrants. There are relatively small CapEx requirements to maintain existing products so pivoting and expanding doesn’t require as much of a operational/financial tradeoff as compared to non-tech. You can keep your previous product running for next to nothing. Plus you can instantly push new changes to existing users.

Software can instantly scale to fill any niche or unlocked growth opportunity. The marginal cost of servicing a new user is basically zero, and it’s instant. Larger companies are held back by old processes, bureaucracy, hyper-focus on a larger revenue stream, etc.

Network effects (which many competing tech companies rely on) can be unraveled. I’ve always had the hardest time arguing this point. Friends have pushed back against me, saying that the whole point of a network effect is that it can’t come undone easily. I wasn’t quite articulate enough to counter that argument, but Marc Andreessen says it well in Elad Gil’s book:

Marc: I think network effects are great, but in a sense they’re a little overrated. The problem with network effects is they unwind just as fast. And so they’re great while they last, but when they reverse, they reverse viciously. Go ask the MySpace guys how their network effect is going. Network effects can create a very strong position, for obvious reasons. But in another sense, it’s a very weak position to be in. Because if it cracks, you just unravel. I always worry when a company thinks the answer is just network effects. How durable are they? To your point on data network effects, I would just say that we don’t see it very often. We see a lot of claims, and very little evidence. The reality is, there’s a lot of data in the world, and a lot of ways to get data. We have not seen very many data moats that actually make sense, even in science.

You can G2M through many channels. Technology companies often go to market through several different hyper-focused channels. All you need to do is find one LTV/CAC positive channel and pour money into it. Non-software companies are often stuck having to do generic lifestyle marketing and branding which doesn’t easily attribute or confirm ROI.

Software is less tied to external forces like regulation, macroeconomic trends, and cultural norms. I’d argue that the whole point of some software companies is to get around or hedge against the effects of these things. Software firms compete more so on product and distribution.

Software is feature-driven. Coke and Pepsi taste the same in blind trials — their slow-changing brands and distribution drive sales. If a software company builds a better feature set, they can win customers on the basis of their product or service’s functionality .

Data is fungible. Users can switch between WordPress and Medium, for example, at will and with no cost. This is a part of what drives software companies to be platforms: if your value is the aggregation and network effect rather than the info itself, users have a cost to switching (fewer eyeballs)

Great software companies tend to be category creators. I disagree with the Thiel view that startups must go zero-to-one or have a grand secret — most amazing businesses can be reframed in incremental terms, like Netflix is just TV but online, or Facebook is MySpace but for your college friends. Most software companies simply don’t compete much on price or quality. To be fair, this point is a bit of a truism: categories tend to be defined by the most prolific companies, not vice-versa.

All I’ve done here is enumerate the some factors in play. The challenging part is figuring out how these different forces work together, in which situations they appear, and what properties of the market emerge from them.

Of course these points are simplified. Many enterprise software companies, for example, have strong lock-in effects on data. It’s very difficult to export your task management or CRM tooling into another system. Perhaps digging more into these exceptions would make for a good followup post.

Agree or disagree with anything written here? Have something to add on? Hit me an email (will@thisdomain) or tweet (@whrobbins)!