I like junk food, and will – from time to time – defend its existence. My defense is this: junk food isn’t 100% bad, for if it were, it wouldn’t exist. People wouldn’t want it.

Junk food isn’t globally bad – it’s bad for you (it will kill you over time), but it tastes good. It doesn’t taste good in the same way good food tastes good. An arugula salad from Blue Plate is good in a way that is different than the way a McD french fry is good. But that junk food definitely tastes good for a real definition of the term “good.”

Otherwise, it wouldn’t exist.

Junk food’s “goodness profile” over time goes something like this: it is good in the moment while you are eating it, but bad almost immediately afterward. You feel sick, fat, gross, guilty.

Healthy eaters are able to train themselves to experience that badness before they make the choice to eat the french fry, and may even experience its taste as a bad taste. (Addicted eaters experience that badness before they make the choice too, but choose to eat the food anyway. That pattern is probably true for all addiction behavior, I’d bet.)

I’ve found this junk food pattern persists across all sorts of human activities.

Writing crappy code even when you could do it right the first time is junk food coding. It feels good to see the software work but you will have to undo all the hax you committed getting there. That will feel bad.

Saying the nice thing instead of the true thing is junk food communication. It feels good to avoid conflict at the time, but you will have to undo the false understanding you’ve created between you and the person you are talking to. That will be bad.

Junk food behavior will kill your company. Here’s the nice thing about junk food behavior though – you know you are engaging in it at the time, and you can correct it. It takes discipline, but it doesn’t take a lot of analysis and self observation. You know it when you are doing it.

Almost no one is able to push themselves to their limits without someone else pushing them too.

Think about it: Athletes all have coaches. Athletes are an obvious example of the raw pursuit of excellence (in generally clearly measurable ways). Great artists often have mentors. Great research groups have great PIs (Principal Investigators), and these groups spawn terrific researchers from their labs.

Even the best, most disciplined people will want to flake every now and then. Even the best, most disciplined people will not always identify how they can improve their process, since they are busy in their process.

That’s why you should get advisors. If you don’t have any, watch out. You’ll likely persist in doing something dumb for too long, or at best, you’ll plateau while your competitors keep ascending. Bad.

Finally, not improving feels bad! Getting better – stretching yourself – is ultimately what you’ll find most rewarding. And it’s really hard to stretch yourself without someone helping you along.

(Thanks to Brandon Roy for stimulating this thought. Not this Brandon Roy, by the way. This one.)

A follow up thought to my last blog post, which suggested that entrepreneurs can be categorized as either artists or strategists:

What is the failure mode for each?

The artist’s company will fail because he finds a local maxima that is not maximal enough. That is, he probably creates a good product, but it may not address a big enough need. He is more likely to arrive at something that works well without thinking about how robust that product’s market could ever be. The artist may also fail because he never ships. His company becomes a historical footnote.

The strategist’s company fails because he doesn’t know how to create a great product. He identifies a need (read: a market), but doesn’t have the taste or skill to create a real, functioning product that delivers in the way a good product does – by being lovingly crafted and obsessed over. His company becomes an also-ran.