There are two models for making a growing, profitable company: the default model, and the scale model.

In the default model, growth is important, but so are margins. Especially if your customers are other businesses, you grow by serving them, but as you attain size you use your power to negotiate higher margins.

In the scale model, it’s unclear if you ever try to increase your margins. For a surprisingly large amount of time, you use your scale to reduce your margins and prices, although you might increase your profits because your scale increases more than your margins decrease.

Some examples:

  • MySpace was default model, with concern for monetization and ads all over its social network.
    • Facebook was scale, with few ads for a very long time.
  • Hosting companies before AWS were default model. Facebook may have been the last tech company to come up before AWS, so built their own data centers. That was just what you did at a certain scale. Of course the hosting company was going to charge you a margin that meant at a certain point it would make sense to bring that in house.
    • But then things changed. Somehow, Amazon started selling S3 and then EC2. It’s remarkable. Netflix switched from self-hosting to using AWS. Now clouds are everywhere.
    • The entirety of Amazon is the scale model. They describe their flywheel as a lower prices -> more customers -> scale -> lower prices.
  • Whatever came before Walmart was default. Become popular -> gain following -> raise prices.
    • I think Walmart might be the first scale company I witnessed. In the late 90s, early 2000s, everyone hated Walmart. They were sure that after they drove out the Mom and Pops, they were going to jack up the prices. But I don’t believe that day ever came. I’m sure they still innovate on logistics and pressure their suppliers to no end. But it seems they have a cultural commitment to lower prices even if they could raise them.
  • Payment companies that services startups were default model. Maybe you used a bank or Paypal, but once you reached a certain scale you would incur the fixed costs to bring payments in house.
    • Stripe is the scale model. How is it that the best way to receive payments for a startup also processes payments for Lyft and Google? Under the default model, Stripe would be good for small and growing companies, but eventually their margins would be too rich, and the larger company would incur the fixed cost to bring it in house. But Stripe is acting like the scale model, promising we will handle this headache for you for such low margins that you will want to keep using it indefinitely.

Here are couple thoughts about the scale model:

  • Patient capital helps. Amazon was unprofitable for a very long time. With lower interest rates going forward, perhaps we can expect more scale companies as well.
  • Customer obsession matters. I have worked at Facebook and Amazon, and both places were suprisingly obsessed with the customer experience.
  • It might not be possible for a luxury brand. Apple is big, but I don’t think they ever tried to dominate a market. At least since the iPhone, they have had fat margins. They do introduce lower-priced models from time to time, but I don’t get the sense they are aiming for market dominance. (Just profit dominance)
  • It will be interesting to see if Tesla can achieve the scale model. They definitely started luxury, although that was a conscious strategy. The question is how far they progress with economies of scale, both in terms of car manufacturing and battery manufacturing. A sign they are pursuing scale would be if other car manufacturers start using their batteries.

I’m not sure what can be learned about pursuing the scale model. And I’m sure there are VC graveyards of companies that took a (realistic or not) shot at a scale model and didn’t make it. I think the main takeaways is that you should be careful shorting a growing company with low or negative margins. They may be playing a different game, which may be higher variance, but isn’t necessarily lower expected value.