To think or not to think

The case for thinking. The case for doing.

I’ve been thinking lately: when you don’t know what to do, sometimes the best thing to do is to stop and think. But at other times, the best thing to do is to take a step. One of the principle components of good decision-making therefore must be to figure out which of these is more important at any given moment. To think or not to think.

I think the way to think about this isn’t so complicated: actions are about getting more data. Thinking is pathfinding.

Source: xkcd

Thinkers vs. doers

I’ve noticed recently that the world tends to divide itself into two camps: thinkers and doers. The doers extoll Silicon Valley hustle culture.

I don’t think either camp is wrong, but in a way, both are wrong. This might sound obvious, but sometimes you have to do things, sometimes you have to think about things. But, unfortunately, this is a sort of naturally polarizing endeavor. You have two options — only one thing will be principally important at any given moment. So you’ll likely learn a lesson at some point that thinking is all that matters (if you don’t think enough) or doing is all that matters (if you don’t do enough).

While this might make sense for you, certainly both are important.

The doing trap

“Think for yourself to decide 1) what you want, 2) what is true, and 3) what you should do to achieve #1 in light of #2. And do that with humility and open-mindedness so that you consider the best thinking available to you.”
- Ray Dalio, Principles

This, especially to you readers of this substack, is probably most familiar to you — I made this substack because I think, in general, people don’t spend enough time thinking, and if you’ve read this far, I suppose you’d agree. In general, I think most important decisions suffer from too much execution, too little thinking. For instance, if your decision will have any scale of impact at all (e.g. if you’re working in a company where your decision will impact the work of many others), the best thing you can do is make certain it’s the right decision. Spending just a few hours thinking about a problem can save you from:

• Losing collective time: Wasting months or years developing a product or running a test that didn’t need to be run.

• Wasting individual time: Everyone below you wasting their time running in the wrong direction.

Good decisions come from good thinking.

The thinking trap

“The best way to see 6 feet further in the fog is to walk forward 6 feet.”
-

Gordon Wong

I’ve found that there are two instances where continued thinking ceases to provide additional value:

• When you reach the limit of your existing data

  There’s a point where incremental thinking is no longer useful, in that you’re extrapolating on thin data. If doing something will get you more data, and if that data would benefit your thinking better than incrementally bashing your head against the wall, it’s a better idea to go get some data.
  
  That said, I think most people tend to underestimate where this point is. As a rule of thumb, for the most consequential decisions, I generally do the following:

  • Spend a couple hours thinking about the problem by myself.

  • Spend a couple hours talking to someone else to get their take.

  • Give myself at least a full day break to let the ideas sit in my head.

  • Spend another couple hours thinking about the problem by myself.

  • Spend another couple hours getting to consensus with the other person.

  
  Usually this means each decision takes at least 2 days to make. There are simply far too many pitfalls in thinking about a decision and making a decision in the same moment — the novelty effect, the validity effect, confirmation bias, etc.

• When the decision is easy to reverse

  I’ll concede that the pattern above is a reflection of my own bias — I’m a founder, so generally the decisions (and the level of decisions I make) are existential. Do we pivot? Do we fundraise? Do we shut down the company entirely? Do we take on debt?
  
  But, of course, many, many decisions are easy to reverse and are not so ambiguous. You’ve likely heard Jeff Bezos’ advice on the matter, which sums this sentiment up quite nicely: when the decision is a one-way door, you need to think methodically, as you’ll have to live with the consequences for some time. When it’s a “two-way door” — an easily reversible decision — the decision should be make quickly.
  
  Methodical decision-making is, of course, slow, and that’s the problem. Setting this as the default for an organization can be crippling — slow experimentation kills the invention flywheel, and this can just as easily kill a company.

In short, better data begets better decisions, and better data usually comes from action.

Final comments

So a final caveat: this is all dependent on so many things — your role, your scope. If you’re an IC and mostly just executing, there’s little need to agonize over your decisions. Usually, the scope is small enough that you ought to just experiment, then move on (well, barring hugely consequential architectural decisions, of course).

But I think this is something folks tend to misunderstand about moving up — you need to make decisions differently. The more senior you get, the more consequential and irreversible your decisions tend to become, meaning the more the tradeoff should tip in the direction of spending more time on the decision.