In Defense of Vertical SaaS

I don’t think AI is going to displace vertical SaaS platforms. It will make the best unbeatable.

At the limit, the vertical SaaS platform teaches and enables the operators to run a specific type of business. The best are created by former operators who became frustrated with the tools available and struck out to make something better. They felt the pain. As a result, they built something intrinsically connected with what they wanted. They had a hard-won opinion about how it should be done. This approach is emblematic of craftsmanship. Or put in the negative, “anti-slop”.

“Slop emerges when we eliminate not just toil (the burdensome aspects of work) but labor itself (the meaningful human engagement with creation).” (1)

The direct connection between the toil and labor of the founder and the usefulness of the product make these vertical specific solutions especially resistant to AI slop disintermediation. And consequentially, difficult to copy or imitate.

So, how does AI change the game for vertical SaaS? Distribution. Operators, when looking for a platform (or to switch) will ask Chat GPT what they should use. Chat GPT will recommend the best solution to them (2), cross referencing the entire internet to help make the decision. Acquisition cost for the best platforms will fall to near zero. They will be able to shift more OpEx to R&D and further press their advantage. Sure, there is probably some basic blocking and tackling in terms of making your website legible to LLMs, but lighting money on fire with Google Adwords, at least in this case, should be over.

In the end, the true craftsmen will win.

Footnotes

  1. Craft is the Antidote to Slop

  2. As LLMs get better, are you sure you’ll be able to fool them with fake reviews? (E.g. LLMs in 2027 - “I went ahead and removed feedback from [username] because their profile wasn’t verified by Reddit”)

Interest Rates & Innovation

I have been aggressively consuming content from folks like Stanley Drukenmiller, Howard Marks, and Bill Gurley as they (and everyone) wrestle with understanding the new normal. While nobody can predict the future (and everyone will be invariably wrong) clear thinking is hard to pass up. Some highlights below.

To understate the obvious, reversing multiple decades of declining interest rates will have systemic impacts that are hard to predict. From Stanley Druckenmiller.

“I've been doing this as some kind of chief investment officer since 1978. And this is about the wildest cocktail I've ever seen in terms of trying to figure out a roadmap”.

The US needs to grow (increase GDP) via continued innovation “fix” our current situation without spurring inflation. In simple terms, we’ve borrowed $31T from the future because we think that what we are working on now will pay off later. Is innovation increasing at a fast enough rate to pay off (1) that debt?

It’s debated (of course). I would consider watching this debate between Peter Thiel and Marc Andressen from 2013. Their main disagreement (from my perspective) is what counts as innovation. Famously, Peter Thiel stated “We wanted flying cars, instead we got 140 characters”. Marc, conversely views innovation more broadly and argues that it's hard to recognize innovation when it occurs. Side note - he created a family of funds to invest in crypto worth $7.6B, so I would argue his perspective is quite broad.

That’s not to say that Marc Andressen hasn’t changed his mind. I read his, much shared at the time, article “It’s Time to Build” as an admission of deceleration.

“You don’t just see this smug complacency, this satisfaction with the status quo and the unwillingness to build, in the pandemic, or in healthcare generally. You see it throughout Western life, and specifically throughout American life.”

Kudos to Marc, this was published before crypto had its monstrous decline. Perhaps he knew Web3 didn’t amount to true innovation – but who wouldn’t give the people what they want for $152M (2) in management fees?

So will we be able to spark innovation? I think so. I remember watching a SpaceX launch attempt on May 27, 2020. It was in the thick of the panic around coronavirus, but we saw calm (3), measured people doing something they felt was important. They didn’t end up launching that day (I believe it was scrubbed for wind or some other reason) but you could tell they felt attached to their mission (4) and accepted the risk. If we all work on things we’re that passionate about, America will be fine.

I wrote this on Jan 26, 2023 with some edits made September 22, 2023.

Footnotes

  1.  Of course we could print $31T and just pay off our debts tomorrow, but can you imagine the price of eggs?? (joking of course)

  2. A16z crypto raised $7.6BM for crypto funds. I assumed 2% management fees – it might even be higher

  3. I wrote this before Walter Issacson’s book came out, Elon appears less “calm” now.

  4. “MAKING HUMANITY MULTIPLANETARY” is one of the best mission statements of all time.