This weekend I built three things. Or rebuilt them. Or more accurately I directed the building of three things while mostly having opinions and doing chores in and around the apartment.
- survey.hd-pt.com, a research survey for my book, collecting data on 75 organizational patterns across six dimensions. Anonymous, five minutes, designed to find correlations between pattern adoption and business performance. (Quick aside here: will you please take this survey, tell your friends to do so, too?)
- ccccharter.com, a better version of a team charter tool I've been wanting for years. New UI, cleaner flow, actually works the way I always wanted it to. Now has organizational support, so you can have a private repo of team charters that everyone can see. Looks great, works great. If you like Glassfrog but don’t like Holacracy, this is for you.
- cpj.fyi/buy, an honestly quite good page to promote Hidden Patterns. Building this was actually the hardest, despite it being the simplest architecturally.
Three functional things, one weekend, with Claude as my teammate. And to echo a thing I saw on Threads, it really did feel like playing a video game. Not in a cheap, gamification sense; there are no points, badges, or leaderboard telling me I'm a 10x developer (I am not devloper). Mostly just that the bad friction was gone. The space between "I want this to exist" and "it exists" collapsed to almost nothing, and I found myself properly working at 8am on a Saturday while making eggs for Emily not because I had to, but because I was having fun.
"I had fun building websites" is not an essay. But I think there's something underneath it that matters, and it connects to stuff we were thinking about fifteen years ago that eventually ends up meaning something to Coase's Law...if you squint.
In 2011, we were deep into collective action design. Mike, Bud and I were working on frameworks for how crowds get motivated to do things together—pulling from Jane McGonigal's game design research, Daren Brabham's work on Threadless, Malone and Laubacher and Dellarocas at MIT. We were trying to understand why some platforms and communities generated huge levels of participation and others didn't, and we had clients that let us apply that thinking to their digital strategy.
McGonigal had this framework from a New Yorker Conference talk, with four reasons people play:
- Accomplish satisfying work
- Spend time with people I like
- Get good at something
- Be part of something bigger
She was talking about games, but to us these were just about motivation. Games just happen to be the best technology humans have ever built for activating all four simultaneously, and that's what makes them sticky. The graphics and the competition are nice, but the fact that you are doing satisfying things, with people you like, while getting better, in service of something that matters...is what matters.
Satisfying work: yes, obviously. Shipping is satisfying. When I asked "can you re-architect this so that it works with Vercel and Supabase instead of Replit" and it just... did it... that's satisfying. The thing I wanted to exist now existed, the way I wanted it to exist.
Spend time with people I like: okay, this one's complicated and probably the most interesting. More on this in a second.
Get good at something: I was learning in real time. I'm not really learning how to code, but at a certain point all of the architecture chat starts making more and more sense. And it changes the way you think about and look at technology—I already knew a lot of the basics but now I really know. This is a skill, and I got better at it over the weekend. Check!
Be part of something bigger: The survey feeds the research. The book site feeds...me? The charter tool serves teams I work with.
Which takes me to Brabham's crowd motivation spectrum with making money on one end and getting addicted on the other end. This weekend wasn't really about making money. Maybe someday I make money off the things from this weekend, but the engagement was all about addiction to my own work.
And now we're back at McGonigal's second point—spending time with people you like. I have a philosophy about collaboration that I've held for a long time, and it goes like this:
The best teammates are the ones whose easy-mode is your hard-mode.
When someone has a skill that is:
- Effortless for them (feels like play, they'd do it for free, it's just how their brain works); and
- Looks like magic to you (you genuinely cannot believe they just did that)
...you're in the sweet spot. The collaboration is pure surplus. You're not competing or second-guessing, you're both a little in awe of each other, and the work is better than either of you could do alone.
The opposite is when you're good at the same thing your teammate is good at. You'd think overlapping expertise would be a strength, but it's often poison. Because when you're also good at the thing they're doing, you have opinions about how they're doing it. You know how you'd do it, and their way isn't your way. Instead of awe, you get criticism. Not because they're wrong, but because you're also right, and that's worse. I've seen this kill projects and teams. Two designers who are both great at interaction design will nitpick each other's work into dust. Two strategists will debate frameworks forever. Two strong writers will edit each other's sentences until everything goes blah.
Respect to my teammates past and present, but Claude Code is the best example of pure teamwork that I've ever experienced, fully recognizing that this sounds insane.
Its easy-mode—writing code, remembering how seventeen different APIs work, debugging, iterating, throwing away an hour of work without complaint when I say "actually, no, start over"—is expansive. Most of the things you can do, it can do as good or better. (I've tried to learn to code probably a dozen times. It doesn't stick. Not because I'm stupid but because it's not how my brain works. Every time I've tried, it feels like homework. Like drudgery. Like the opposite of play. npm install what the hell are you talking about) You constantly feel amazing because you're constantly making progress. The outputs just keep getting better.
My easy-mode? It does not give a shit about my easy mode. It does not notice whether I'm good at having opinions about what should exist, knowing what good feels like without being able to articulate the rules, understanding organizational problems deeply enough to build tools that actually address them, or even having taste in product. It does not care and will build you anything, beautifully, and have absolutely no idea whether it was worth building. And the AI never gets tired. Never gets defensive. Never says "well I liked the first version better" in a tone that means something else entirely. It just... builds. And rebuilds. And rebuilds again. At 2am on a Sunday, it has exactly as much energy as it did at 9am. I realize that it's peak hype cycle to talk about AI not getting tired, and not discounting its sycophantic behavior, but when you're making a thing, this sorta matters.
Good collaboration is mostly just being down
Bigger picture: SaaS appears to be over. The market dip this week, it seems, was at least partly a reaction to Claude Code. If one person with an AI can build a charter tool, a survey platform, and a book promotion site in a weekend, what does that mean for the companies charging $29/month for each of those things separately? A lot of mid-tier SaaS exists not because the idea is hard, but because the building was hard. The value proposition was: "we did the annoying technical thing so you don't have to." When the annoying technical thing takes a weekend instead of a year, that value proposition evaporates.
This is basically Coase's Law playing out in real time—when the cost of building drops below the cost of buying, the buy market collapses.
I'd expect that anything that was a wrapper around effort is toast. The stuff with real network effects, proprietary data, irreplaceable expertise (What counts here? Figma? Notion? Akamai?) probably lives on much longer. But "nice form builder" might not, when making a nice form builder is a Saturday afternoon.
I want to be careful here because "work should be fun" is a sentence that's easy to agree with and equally easy to misunderstand. Not fun like a pizza party, or like a ping-pong table in the break room. I mean fun like: the thing you're doing is intrinsically satisfying, you're getting better at it, you can see the results, and the friction between your intention and the outcome is low enough that you stay in flow.
That's what games get right! It's the feel of an immediate feedback loop. Do a thing, see the result, adjust, do it again, lose track of time. Most work isn't like this. Most work is: you have an idea, you write a brief, you wait for someone to read the brief, they misunderstand the brief, you have a meeting about the brief, you revise the brief, it goes into a backlog, it gets deprioritized, it comes back three sprints later looking nothing like what you wanted.
I don't know what this means for individual jobs or companies, but I do know that this is going to spread to more and more people and when it does, friction disappears, motivation changes, and what's possible feels completely different from one day to the next.
A few things are true simultaneously:
- I've had some very fun jobs and this weekend was very fun. The things I built are good. Not "good for AI-generated." Good good. I would put them in front of clients and feel proud.
- Something is also being lost, or at least rearranged, and I'm not sure what or how to think about it. Is the struggle part of the value? Is human-made important? For what categories of things, and why?
- What happens when the hard parts that are fun for some people just aren't hard anymore? My fun weekend is someone's displaced livelihood. I don't have a good answer for this, and I'm suspicious of anyone who does.
- Is "it feels like a video game" a warning sign? Games are also addictive. They're also designed to exploit dopamine loops. They're also, sometimes, a way to feel productive without actually producing anything that matters. I don't think that's what happened this weekend, but the feeling of flow can be a lie. You can be in flow doing something pointless.
Anyway. Everything probably changed in the last few months. For the better or for the worse, I'm not totally sure. I do know that it is really fun, and I'm going to go build more things that I've wanted to see in the world until I know which end is up.