← All issues

The Compound Years · Issue #13 · August 2026

The cost of "I'll do it later"

Most ”later” projects take a weekend if you start now -- and a decade if you do not.

01The experiment

Built a working app in a single weekend instead of a six-month plan.

I had been ”planning” a side project for months. One weekend I sat down, scoped it to what I could finish by Sunday night, and shipped a working version. The plan I had been writing for months was about ten percent of what I actually built when I just started.

02The lesson

Later is a story you tell yourself; the dataset comes from starting now.

I have a folder of ”later” projects from three years ago that I genuinely still have not touched. The ones I started inside a weekend taught me far more, even when they failed, because doing the thing once is worth more than thinking about it ten times.

03The founder

Drew Houston shot a Dropbox demo video before the product existed in 2007.

Houston could not finish the product fast enough to pitch it, so he recorded a three-minute demo, posted it on Hacker News, and watched the waiting list jump from five thousand to seventy-five thousand overnight. The video was the experiment -- and it proved demand before he wrote production code.

04The mindset

The cost of ”later” is invisible until you add it up.

Every ”later” I keep on the list takes up some background tab in my head, slowing everything else down. The habit I am building is to either start it this week, schedule a real day for it, or honestly delete it from the list. Anything else is just background noise.

05The question

What is one ”later” project you could actually start this weekend?

Hit reply and tell me. I read every one.


Forward to one person whose ”later” list is older than your relationship with them.

-- David