How much do you need to understand?

One of the many traits I admired about my dad was his relentless desire to understand why things were the way that they were. 

As a software engineer, he would rather pour over documentation to solve his coding problems than copy and paste StackOverflow code that he didn’t fully understand. 

When I needed his tutelage for math assignments, he insisted that we reason our way to the solution instead of relying on mindlessly copying and applying the suggested algorithms for the textbook. 

In reflecting more broadly on the implications of this character trait, it occurred to me that having superficial knowledge —the mere awareness of what works without understanding exactly why — is probably sufficient for about 90% of the problems we’re faced with. 

If it works, it works. 

It’s only in the remaining 10% of problems — typically ones that require a high degree of creativity or optimization — that the benefits of having a clear and nuanced understanding of why things work come to the fore. 

As a data scientist, your algorithms can only be so accurate before you need to apply specialized knowledge about parameter tuning. 

As a strategist, your solutions can only be so effective before you need to start applying specialized knowledge about economic theory, market dynamics, and consumer psychology. 

As with most things, there is a balance to be struck when it comes to figuring out how much knowledge you really need.

In general, having a deep understanding is invaluable for work where optimization is vital. 

In all other contexts where the need for functionality far supersedes the need for optimization, you can save your energy — your pursuit of knowledge — for other things.