Some Notes on Moats
Moats, RIP Bowery Farming, election reaction
Moats are important. Duh. Economic moats drive long term margins and profits for any business that gets big enough to attract jealousy and competition.
Startups don’t have moats. If they did they wouldn’t be able to start up in the first place (with very few exceptions in extremely technical or regulated industries). But every founder should be thinking about the eventual moat from inception.
There’s a few questions to ask:
When will your moat kick in? Self explanatory.
How much of a moat will you have/how strong will it be? This can vary greatly.
What kind of moat will you have? Could you have more than one?
There are only a few moats that count. Lmk if you think i’ve missed something.
Network effects or demand side economies of scale. The service gets better with more users. These can be stronger/global (Facebook) or weaker/local (Uber). Sometimes this is derived from unique data, sometimes from connectivity, sometimes from
Economies of scale. You get better pricing with size (Walmart) and/or can amortize capital investments over a big base that others can't (Buffett's local newspaper thesis). The other side of network effects
Regulatory. The government has picked you as a winner either explicitly (Boeing, x-Vegas casinos) or implicitly (any licensed FI)
Intellectual property. You have strong patents (biotech) or it's just near-impossible to replicate your technology (TSMC).
Brand. A close cousin of IP moats. People just love you for ineffable even if irrefutable reasons (McDonald's, Coke).
Not included here are switching costs. Generally that defends customer relationships rather than defending the market/opportunity. Switching costs are only a moat for the biggest companies in low growth categories and are therefore irrelevant to startups.
Other things that aren’t moats or are already covered here
Capital. “Capital is a moat” is a very 2021 thing to say. Google can pay Apple for placement because it’s profitable elsewhere because of a real moat. Arguably Google can pay because of an implicit regulatory moat that is now being challenged in federal court. That is anyone can pay; only Google can get away with it. Even if capital were a moat (which it’s not), it would be the least relevant for startups - no amount of money you can raise and spend is even relevant next to real companies.
Distribution. This sounds like a moat but it’s not unique among the list I set out. A “distribution moat” is almost always an economy of scale story (supply or demand side), a brand story, or just more of an advantage than anything super structural. For example, Feastables doesn’t have a distribution moat - Mr Beast has a brand. The moat belongs to him, not the business (which is why Beast rather than Feastables is worth something).
If you have an example that breaks this logic, chuck it at me.
So here's the kicker: you won't have a moat for a long time. You might never even have a real moat - pretty much no SaaS companies do! That means that in the short-medium term you just simply have to out execute people before you can out earn them.
And execution ability doesn't come through from a resume or a deck so it's hard to back early without spending a lot of time together. But if you want to build a really killer business, think about moats before you get started.
h/t to Sam Gerstenzang for the back and forth to sharpen this up.
On The Election
As any of you who know me well would know, I am pretty devastated about the outcome of the election. I think Trump will be a disaster for the country and the planet. Last time was bad, this time will be worse. However I’m mindful that none of you follow me for my politics and it’s not my area of particular expertise.
So I plan to generally (but not exclusively) stick to the promise I’ve made and deliver on what I think you’ve all signed up: takes on startups, venture, business, etc. Tl;dr I reserve the right to talk/write about politics when I feel so moved but am aware that no one cares what I have to say.
I wrote
There are only 5 moats. Which one will you have?
I read
Almost nothing. Terrible information environment.
I watched
Twin peaks. Getting into it for the first time.
This was a lightbulb moment for me:
"Startups don’t have moats. If they did they wouldn’t be able to start up in the first place (with very few exceptions in extremely technical or regulated industries). But every founder should be thinking about the eventual moat from inception."
Would you agree that an exception for network effects is a middleman/gig economy platform model like Uber and Airbnb. They need network effects from the beginning? Or else they can't take off.