DHH: How to Build a Profitable Company Without Losing Control

0:07:43 · watch on youtube.com↗

Our reading

The competitive threat to a small software company is a team of two or four working under the same constraints, never the behemoth.

Their wordsI'm worried, and still am, about a team of two, about a team of four, because they will produce software that's competitive with ours because it comes out of the same constraints. And therefore, our threat doesn't come from the behemoth.

0:09:00 · watch on youtube.com↗

Our reading

AI acceleration threatens 'less software': without the constraint of scarce hours, even 37signals will likely build too much.

Their wordsSo, this is why I'm quite skeptical about all this new AI acceleration we're having is even us, I think it's quite likely that we could use it wrong to build too much software, to get too far away from less software, learnable software. It still has to fit in the mind of a human.

0:12:16 · watch on youtube.com↗

Our reading

Venture money ruins software: raise a fortune, hire hundreds, and crap comes out, because the funnel that forces hard choices is gone.

Their wordsThis was one of the reasons why I was always so skeptical about venture capital when it came to software. Because I saw time and again people raise an enormous amount of money, hire hundreds of people, and produce crap software.

0:18:18 · watch on youtube.com↗

Our reading

Old versions are kept until the end of the internet: Basecamp's 2004 version, unsold since 2010, still makes millions in profit.

Their wordsSo, for Basecamp, we kept those old versions. The version we launched in 2004, we stopped selling in 2010. 16 years ago. We still have customers on it. As an individual business, it's still making millions of dollars.

0:20:32 · watch on youtube.com↗

Our reading

Software should be allowed to be finished, the way a knife or an album is, and the customer who bought the old version should get to keep it.

Their wordsAnd I actually wish we had more of a capacity in software to accept finished products. This is one of my great envies of people who make physical products. By the time you ship, I don't know, a knife, it's done.

0:23:41 · watch on youtube.com↗

Our reading

His career only took off once he could close a door and see no one for long stretches; open offices flatlined his productivity and creativity.

Their wordsAnd in fact, when I look back upon my career, it only took off when I could close my door and not see anyone for long stretches of uninterrupted time.

0:32:29 · watch on youtube.com↗

Our reading

He was wrong to lock in on autocomplete as what AI would be; Tobi Lütke saw further ahead, and he is frustrated he lacked that conviction.

Their wordsSo, I didn't like that at all. And I therefore locked in on thinking this is what AI is going to be. And Tobi clearly saw much further ahead.

0:36:17 · watch on youtube.com↗

Our reading

Opinions move by touching things, not by reading about them: nobody learns to drive a race car from a book.

Their wordsAnd so much of the ability to move your opinion comes from not these abstract ideas, but from touching things, from actually setting up open claw or whatever and getting some experience with it and then having this instinctual feeling of where things are going to go that you're not going to get from just reading someone else do it.

1:02:28 · watch on youtube.com↗

Our reading

The ideas that end up mattering sound stupid at first, because that is where the value is locked up; copying the best with a tweak never works.

Their wordsAll the best ideas that actually turn out to have an impact on something, they sound stupid at first. They're not intuitive because that's where all the value has been locked up.

1:07:42 · watch on youtube.com↗

Our reading

Out-teaching the competition built 37signals for two decades, and algorithmic feeds are now breaking the reciprocal exchange it relied on.

Their wordsWhat's broken now on the social platforms is that my followers don't see anything I post unless it goes medium viral. Because everyone is now in the algorithmic feed.

1:13:04 · watch on youtube.com↗

Our reading

Five-minute feedback on a 4,000-hour gift is a privilege you pay for: customers earn it, free users owe a modicum of respect first.

Their wordsYou do owe me just a modicum of respect to look at this more than 5 minutes before you just dump your shitty ideas onto me or even worse, it's not so much about the ideas, it's when people try to tell me what to do.

1:14:53 · watch on youtube.com↗

Our reading

Retirement, when it comes, will be a retirement from customers and commerce, not from building and sharing things.

Their wordsBut I've also realized at some point when I retire, it will actually be a retirement from capitalism. It'll be a retirement from customers and the economic exchange.

1:17:55 · watch on youtube.com↗

Our reading

Someone who makes a lot of money selling something good has a moral obligation to buy the Ferrari: it keeps the craftsmen who make beautiful things in business.

Their wordsWe would not have Ferraris to pine for if no one bought them. And who can buy them? The people who've been successful in other parts of the business. So therefore, it is your moral obligation to buy a Ferrari if you do really well.

1:23:54 · watch on youtube.com↗

Our reading

Revenue is outside your control and costs are inside it, which is why cutting a recurring expense is the lever a business owner should love.

Their wordsBut ultimately, I don't get to choose whether they put in their credit card or not. They do. I do get to choose how much money we spend on servers. I do get to choose whether we squander that revenue on needless cloud services rather than just owning our own hardware, right?