Each week I split test my Founder Finds newsletter. One change per test.
- A word
- A number
- The order of a phrase
I’ve run over 150 of them.
At some point it became hard to find new things to test. I’d swap something and the numbers held steady.
That was the signal.
The newsletter had become hard to vary.
David Deutsch coined this in The Beginning of Infinity. A good explanation is one where changing any part makes it worse. Remove a piece and it breaks. Swap a variable and it falls apart. Each part is load-bearing.
The paperclip has looked the same for over a century. Dozens of redesigns have been patented. The original still wins.
The same logic applies to things you build. A skill, a product, a piece of writing. When you can’t change it without degrading it, you’ve found the shape.
I wrote about starting at the skateboard. Ship rough. Iterate fast.
This is about when to stop.
You stop when edits vanish.
Hard to vary is rare. Most things keep changing.
📎
What have you been refining that might already be done?
