Find a model

One of the most common pieces of advice I give to mentees is to "find a model." It's a stupid-simple technique, but can transform how you work. It will let you do things faster, better, and with more confidence. It's a mystery that everybody doesn't use it all the time.

The core of the technique is to find a model artifact to keep an eye on while you work on your artifact, which could be a coding project, a song, a painting, an email, a social media post, an essay like this one, etc.

The are three important benefits to this technique.

A model gives you a north star; something to shoot to; something you can visually compare your work to, and see what's missing; a yardstick by which you can measure how "done" your artifact is. Without a model, how do you even know what kind of thing you're making? How can you tell its shape? A model is like a hyper-specific template that lets you stay focused on the end goal.

There are many related principles: Backward Chaining, Top-down design, The XY Problem, and Amazon's principle to "Write the Press Release First". These all get at the idea that you don't want to lose sight of the end goal. Without a model, you risk getting lost in the details. Losing the forest for the trees. A model helps you keep the whole forest in sight, so you actually create what you set out to.

However, it's important to remember that "Plans are useless, but planning is indispensable." In other words, you shouldn't shouldn't feel compelled to stick to your model so closely. Use the parts that you want, and forget about the rest. A model is high up on The Ladder of Abstraction, but that doesn't mean you can't change it. If you realize that you chose the wrong model or that your goals have changed, excellent. Time to find a better model.

The fact that you have a model is what allows you to realize when you have the wrong model. Imagine you're a ship captain navigating by a star that you thought was the north star, but then later realized was just Venus. The time you wasted navigating by Venus would've been equally wasted if you were navigating without reference to any star. At least when you're navigating by Venus one of your fellow sailors can come over your shoulder and say, "Hey captain, I think that's Venus." If you're navigating by nothing, then you're truly lost at sea.

Another reason to have model is that it's very helpful when you get stuck. Don't know how to do something? See how your model does it.

For this essay I felt stuck on the first line. This essay's model is Do things that don't scale, because it's an essay that explains a cognitive technique. Its first sentence isn't the prettiest, but it gets the idea across and it's a hook: "One of the most common types of advice we give at Y Combinator is to do things that don't scale."

Now go back and read my first line. Is it deritive? Yes. Would literally anyone have noticed if I didn't draw your attention to it? No. There's something profoundly true about the line oft-attributed to Picasso, "Good artists borrow. Great artists steal." I didn't truly believe this quote until I read The Elements of Eloquence, wherein he shows examples of Shakespeare lifting whole sentences out of other books and plays, only changing a couple words.

In school, they call this behavior cheating, which is probably why this technique isn't practiced more often. I think it's an important lesson to unlearn. Why reinvent the wheel? Save time so you can focus on your unique contribution. Just be sure to give attribution and credit where it's due.

How many times have you built something only later to realize someone else already built it? Chances are that if an idea occurs to you, likely it has occured to someone like you. So many people don't even take a second to google around before embarking on a project. Instead of wasting your time reinventing the wheel, see if someone already invented it. Then stop. Just use their thing.

One reason "finding a model" isn't more common is that it can be quite difficult to find a good model. I call that technique "surveying the space".