10-90 Problems

I have lots of things I’d like to write that seem important, but I can’t seem to write any of them, so I’m writing up this dumb small idea in order to get momentum.

80-20 problems are where 20% of the effort gets you 80% of the way there. They imply an effort-reward curve like this:

a curve that goes up sharply until you get to about 20% effort/80% reward, then gradually up to 100/100

One example might be cleaning the house. You can deep clean every square inch (100% effort) or you can just put in about 20% effort and it will probably feel 80% clean.

A lot of problems are a similar but different reward curve. I’m going to call these 10-90 problems1. The reward curve for them looks like this:

a curve that goes up sharply until you get to about 10% effort/50% reward, then stays flat until 90% effort/50% reward, then goes up to 100

You get half the payoff for doing the first 10% of the effort, and half the payoff for the last 10%, and no payoff in between. An example might be “being well dressed.” If you put in zero effort, people will think you look like a slob. If you put in just 10% effort, people will not think twice about how you look. And this persists until you get to about 89% effort! It’s hard to look stylish enough for people to notice you. To actually look particularly well-dressed, you have to be in the top 10% of effort.

Why is it like this? Maybe because “being well dressed” is actually two goals: 1. not to be noticed as bad, and 2. to be noticed as good. Or maybe it’s because “being well dressed” has a plateau in it - perhaps because it’s related to some kind of bell curve. (being within 2 SD of the mean is easy - but being a positive outlier is hard.

So what? It’s just good to notice when you’re in a 10-90 problem. Often your instinct is to put 50% of effort into something - but any amount of effort from 11% to 89% makes no sense! Go for 10%, or go for 100%.


  1. even though that’s a comically bad name! it doesn’t even contrast memorably with “80-20”! but it makes sense in my head and I’m writing a dumb small idea to get momentum. ↩︎


blog 2024 2023 2022 2021 2020 2019 2018 2017 2016 2015 2014 2013 2012 2011 2010