Nobody owns the small stuff

I’m building 52 internal tools in 52 weeks. Not as a portfolio stunt, but to test a belief: most business inefficiencies don’t come from a lack of engineering resources. They come from missing small, high-leverage tools that nobody is responsible for building.

Every company bleeds money through tiny inefficiencies. Requesting time off through two separate platforms. Product announcements that take a week of back-and-forth to publish. The problems are visible to everyone and owned by no one, because there is no role whose job is the small stuff. Engineering has a roadmap. Vendors want six-figure contracts. The small problems just sit there, compounding.

Here’s the arbitrage: a tool built in days costs a fraction of what a vendor solution does, not because the builder is cheaper, but because the problem never needed a vendor in the first place. Week two of the project made the point for me. I heard colleagues frustrated about publishing announcements during a morning meeting, and by evening I had shipped Megaphone, a self-service publishing platform with tag-based targeting and AI writing assistance. Seven hours from complaint to working software.

That speed is new, and it changes what a designer is. I used to sketch on paper to think through an idea. Now I sketch in an LLM. Describe the flow to Claude Code, get an interface I can actually click, and find out in minutes whether the idea survives contact with reality. Working software has become the new scratch paper: cheap enough to throw away, real enough to learn from. Which means design no longer has to wait for a build team to find out whether it’s right. You can listen to a pain point in a meeting and put something testable in front of people the same day.

The lesson from the first five tools is consistent. Speed beats perfection: functional today does more than polished later. Small tools face less resistance than ambitious ones. And the gap between “person who notices the problem” and “person who can ship the fix” has collapsed. You no longer need to be an engineer to ship real software. You need to listen well, build fast, and know what good feels like when you’re holding it.

Somebody should own the small stuff. I’m making the case that it should be a designer.

Adapted from a longer essay at One Tool a Week.