Premature Optimization is the Root of All Evil

"We should forget about small efficienceis, say about 97% of the time: premature optimization is the root of all evil" - Donald Knuth
H/T: @lpolovets

Donald Knuth was specifically talking about algorithms in computer software, but the lesson applies broadly to product development, startups, and self-improvement.

Perfection is the Enemy of Progress

Wikipedias worth of man-hours have been spent building features that the customer never requested and rarely use. Companies have spent millions on inventory to achieve economies of scale on products that crater in the market. They should have shipped first and asked questions later.

Stop futzing around and ship it! Premature optimization is often procrastination in disguise. If you’re “making it better” before someone has used it, you’re letting fear of judgement keep you from learning.

Don’t worry about scaling, don’t worry about “nice to have”, just start! Start going to the gym, spearhead a new process at work, ask someone to buy your barely functional prototype. Even if you fail, you’ll be at the same place as if you were still planning; the only difference is, you’ll have a data point from the real world. Now you can adjust, incrementally better, then ship it again.

Questions of the Day

What aspect of your life should you be shipping instead of optimizing? Is there a feature in your product or startup that’s a good idea but you haven’t had a customer ask for it yet? Tell me about the time you sunk days of oyur life into solving a problem that didn’t exist.

Leave a Reply