Ten principles
Written in January 2026, before Loop shipped. Kept as it was, as a record of what the studio thought it was setting out to do.
Written in January 2026, before Loop shipped, and kept here as it was. We have since come to prefer letting the tools speak. This stays as a dated record of what we thought we were setting out to do.
These are the principles we return to when we’re stuck, when we disagree, when the easy path and the right path diverge. They’re not rules. They’re compass headings.
I. Respect attention
Attention is the most valuable thing a person can give you. Treat it accordingly. Never take more than you need. Never manufacture urgency. Never exploit psychology for engagement.
II. Ship less
The feature you don’t build can’t break, can’t confuse, can’t bloat. Every addition should justify its existence against the alternative of nothing. Most of the time, nothing wins.
III. Charge honestly
If the product is free, the user is the product. We charge money because it aligns our incentives with our users’. We succeed when they’re happy, not when they’re addicted.
IV. Design for the worst moment
Don’t design for the user who’s alert, focused, and patient. Design for the user who’s exhausted, anxious, and has three seconds of willpower left. If it works for them, it works for everyone.
V. No dark patterns
No trick questions in modals. No pre-checked boxes. No “Are you sure you want to miss out?” copy. No hiding the unsubscribe button. No guilt. No manipulation. Ever.
VI. Own the whole stack
Understand every layer of what you build. Don’t outsource your core to a library you haven’t read. Dependencies are fine. Ignorance isn’t.
VII. Prefer deletion to addition
When something isn’t working, the first instinct should be to remove, not to add. Complexity is a ratchet. It only goes one direction unless you actively fight it.
Perfection is achieved, not when there is nothing more to add, but when there is nothing left to take away.
Antoine de Saint-Exupéry
VIII. Test on yourself
Use your own products. Daily. In real conditions. Not in a demo environment with perfect data and good lighting. Use them tired, distracted, in bad lighting, on a slow connection. That’s where the truth lives.
IX. Write it down
If a decision isn’t documented, it wasn’t made. It was assumed. Assumptions rot. Write down why you made a choice, not just what you chose. Your future self will thank you.
X. Stay small
Growth for growth’s sake is the ideology of the cancer cell. Stay small enough that every person on the team knows every product, every user complaint, every line of code. Scale the impact, not the headcount.
These principles aren’t original. Most good ideas aren’t. But writing them down, committing to them publicly, makes them harder to quietly abandon when the pressure mounts.
And the pressure always mounts.