So… I have a problem with building things.
I (like most of you) now have the power and tools to take any idea and start building it in about ten minutes. But see, even though it feels like ten minutes, is it ever actually ten minutes? No, nothing is ever ten minutes. It’s ten minutes to the first working version, and then it’s the rest of my week perfecting it.
Building stuff is the new doom scrolling for me. Just one more prompt. One more little fix. One more thing that would make it better. I know a handful of other people who go head first into this the same way I do, and we all have the same look in our eyes.
I timed myself and gave myself about five minutes to think through all the stuff I’ve built this year:
- A whole second brain to run my life.
- An email tool that scanned my inbox and drafted replies
- A home app so my wife and I could stop living inside Apple Notes.
- A financial dashboard for the company that I then tacked on: event management, a CRM, calendaring, to-dos, and proposals.
- A full system for building WordPress sites so we’d quit rebuilding the same parts over and over.
- A sitemap tool to triage a 500-page website. Then a SECOND sitemap tool for the next project, because obviously I couldn’t just reuse the first one.
- A client onboarding studio built on top of that second sitemap tool (…whicih included a third sitemapping tool).
- Workbench, where I keep everything I’m writing.
- An events app I built with my buddy Mickey after our Google Sheet turned into a nightmare.
- A foundation system to stitch all these mismatched apps together.
- A podcast pipeline that uploaded and organized episodes for me automatically. Worked beautifully. I quit podcasting right about when I finished it.
- A CMS built on top of my style guide, for a friend’s idea we might never run.
- A stripped-down D&D character sheet, because the popular one drives me up a wall.
- And, on the topic of D&D, a full D&D campaign platform. Regions, towns, characters, session notes, and an MCP so the whole table could stay on the same page.
Hundreds of hours and most of it I don’t use.
Yes, this took from the business as every distraction does. We get pulled toward the thing that’s more fun, and building is WAY more fun than the boring, important work. Nobody ever got distracted from their sales follow-ups by doing more sales follow-ups. Don’t get me wrong: you feel great after you finally do the boring stuff, but you never crave it (or at least I don’t).
This is the reason I’m writing any of this down.
I don’t think of these as thrown away. I don’t delete the code and I have no plans to do so. There’s zero embarrassment about spending thirty hours on something I’ll never ship. A little regret, sure, but only on the builds that were pure dopamine and nothing else. The rest taught me something.
All of this is a new language. Everybody right now is learning to talk to the robots. What they like, what they don’t, how to be specific enough but also loose enough (because sometimes they do better with less). How to manage context. How to hand work off to sub-agents. How to actually QA the thing. You don’t learn a language by reading about it. You go live in it and you speak it badly for a while until one day you don’t.
The initial sitemap tool I built showed me that it in some cases, it’s actually simpler to build a mini-app than to use traditional tools — even for something that is a one-ff (which, btw, the client loved it). The one I use now is the third version, and it’s fantastic. An AI can talk straight to it. It can draft a sitemap from a client’s live website. I can drag and drop the whole thing around live on a discovery call while the client watches it happen. That version would never have been made if I didn’t “waste my time” on the first two.
But — some of them stuck. The operations dashboard is the one I always come back to. I used to create new spreadsheets year-over- year to track my finances because there is always a better way to do things. Granted, I’ve only been using this new-n-shiny system for a few months now, but it has all the bells-n-whistles I’ve ever wanted in a dashboard application.
The events app with Mickey went from a broken Google Sheet on a Tuesday to a full working app by Sunday. Now we have 12 events planned out for the rest of 2026, a real list of people with their photos and emails and history, and calendar invites that I started sending out yesterday. That should not be possible in five days. The whole point of that app is connection, and it’s helping me do that.
I’ve noticed that the bigger I make an app, the less I use it. The events tool lived inside my ops dashboard first and I never touched it there. I pulled it out into its own small single-purpose app, and now I’m in it constantly. At its core it’s a me problem. But … looking at it from a different angle, it also makes total sense, because a tool built to do one thing has a cleaner face and gets out of your way. A tool built to do everything gives you the most generic experience on earth.
So… what’s the actual rule going forward to not chase EVERY idea that I have in my brain? Because simply saying to myself “you should have more self-control” is a wish, not a rule.
I’m leaning towards this: I only get to build it if it makes a real person’s life easier, and that person isn’t me. My own life is pretty simple at this point. The mess starts when I build a new thing to manage my things. The second brain is the cleanest example I’ve got. Instead of just answering my email, I built a system to answer my email, which needed my transcripts, and automations, and it needed to learn my voice/tone, and on and on it went. The tool I built to kill the noise became the noise.
Will I hold to that rule? Psh, I don’t know. I know my own personality VERY well. But it’s a good place to start.
One more thing, because I don’t want you to read all of this and hear “your building was pointless.” That is the exact opposite of my point.
If you’re a founder spinning your wheels on this stuff, especially if you don’t come from design or development, keep going. A lot of you are making genuinely cool things. And, there’s a magic in NOT knowing what I know. I put myself in a box because I remember how much it used to hurt to wire one tool into another. Because of that, I hesitate before I even try. You folks don’t have that baggage. You write a couple of sentences and the thing that used to take me a week is just done.
So go build and just try new stuff. It’s how we learn. Just don’t chase every single idea that wanders through your head at 10pm, because not every idea is a good one, and most of them will never move the needle.
My goal here is to walk through these apps one at a time. What each one is, why I built it, how it’s actually put together, and anything I learned from the building process and/or what I learned from it sitting on the shelf.