My whole system of life is keeping at it. The task of life is not to see clearly in the distance but to do the task at hand.
- Charlie Munger
Let's talk about planning and prioritization.
- People who trust big, fancy Gantt charts are destined to be disappointed. Why? Because anything beyond the short-term is too volatile for rigid planning to be useful.
- If you're unwilling or unable to quickly change the plans as needed, that means you're not adaptable. Eventually, you'll end up working on the wrong thing. Arguably, that's worse than doing nothing!
- An inability to account for (and adapt to) real-life volatility is a sign of bad planning. A plan that assumes a perfect world is a bad plan.
Well, that's just fantastic. Should we scrap all our plans and yolo it? Not quite. Enter the "distributed Pez dispenser" model of planning. It's basically gradient descent:
- Gather everything you may want to work on in the foreseeable future.
- Put it in a fuzzy priority queue. This is our "distributed Pez dispenser."
- Pop from the top of this Pez dispenser. The "next" item to work should be clear. It's the most important thing at the given moment. So focus on that. Do it well.
- Add to, consume from, or rebuild the Pez dispenser as circumstances shift.
Properties of the Pez Dispenser
The rest of the contents of the Pez dispenser should be a "gray box." That means you don't really know its exact contents, but you should have a rough idea. A "rough idea" means you have two things:
- A north star of what must be done.
- A general sense of what may change.
You don't need to over-index on anything beyond what's next, because it's subject to change. Actually, it's unlikely that it won't change, given the volatility of real life. Notice how this model accounts for such volatility while allowing us to make incremental progress. The next step should always be clear, and you should never be married to anything after that.
This concept applies to life in general as well. Plans change, but that doesn't mean you don't make plans. Rather, avoid rigid plans. Have adjustable plans, and execute the current step well. Then rinse and repeat.