How I Decide What to Work On

The Observation

There’s no shortage of interesting work. New projects, new opportunities, new problems to solve. Each one seems worthwhile on its own merits. The list grows faster than any one person could work through.

This abundance creates its own problem: how do you decide what actually deserves attention?

I’ve tried various approaches over the years. Most of them failed in predictable ways. What follows isn’t a system that works for everyone—it’s a framework that works for me, refined through repeated failures.

What Breaks Without Constraints

When everything interesting is treated as worth pursuing, specific failures emerge:

Scattered attention. Time fragments across too many things. Each project gets enough attention to stay alive but not enough to reach completion. The result is a portfolio of half-finished work and no meaningful outcomes.

Context switching costs. Moving between projects isn’t free. Each switch requires reloading context, re-establishing momentum, re-engaging with the problem. Frequent switching means most time goes to transitions, not progress.

Diminishing returns per project. When you can’t give sustained attention, you can only work on easy parts. The hard problems—the ones that actually matter—require focus that fragmented attention can’t provide.

Opportunity cost blindness. Saying yes to one thing means saying no to everything else, but this isn’t visible when decisions happen individually. Each yes seems reasonable; the accumulated yeses create impossibility.

Completion failure. Nothing finishes. Projects stall at 80%, waiting for the final push that never comes because attention has already moved to the next interesting thing.

The Filters I Use

Not every question needs answering. Here are the filters that survive:

Does it matter if I don’t do it?

Many interesting problems will get solved by someone else if I ignore them. Many interesting opportunities will recur if I pass this time. The question isn’t “is this worthwhile” but “is my involvement necessary?”

If the answer is no—if the world moves forward fine without my participation—that’s a strong signal to pass.

Does it compound?

Some work produces one-time value. Some work builds on itself. An article published creates permanent value and can attract opportunities indefinitely. A one-off consulting engagement produces value only during the engagement.

I bias heavily toward work that compounds: systems over hours, assets over activities, infrastructure over events.

Am I the right person for this phase?

Projects have different phases that require different capabilities. Some need early-stage creativity. Some need execution discipline. Some need refinement and polish.

I’m better at some phases than others. When a project needs capabilities I’m weak in, my involvement might actually slow it down. Being realistic about fit matters more than enthusiasm about the topic.

What’s the failure mode?

Every commitment has a failure mode—the way it will go wrong if reality diverges from expectations. Understanding this in advance changes the decision.

Some failure modes are recoverable: you can stop, adjust, try again. Some failure modes are catastrophic: the failure damages other things you care about. I avoid catastrophic failure modes aggressively, even when the expected outcome looks attractive.

What does this displace?

The commitment doesn’t exist in isolation. It takes time and attention from other possibilities. What specifically am I not doing to do this?

Sometimes the displacement is acceptable. Sometimes, when I actually list what gets displaced, the trade looks worse than it seemed initially.

The Decision Pattern

When something new appears that seems worth doing, I run it through a process:

1. Wait. Not immediately, not reactively. Let it sit for at least a few days. If it still seems important after the urgency fades, it might actually be important.

2. Name the cost. What will this actually take? Not the optimistic estimate—the realistic one. Include context-switching costs, opportunity costs, and the cost of potential failure.

3. Name the displacement. What specific things will not happen because of this? List them explicitly.

4. Check for ego. Am I attracted to this because it’s genuinely valuable or because it’s flattering to be asked, interesting to think about, or exciting to tell people about? These are different things.

5. Decide once. If the answer is yes, commit fully. If the answer is no, decline clearly and don’t revisit unless something material changes.

The Default Is No

The most useful change I’ve made is shifting the default.

Previously: new opportunities started as maybes that needed reasons to become nos.

Now: new opportunities start as nos that need compelling reasons to become yeses.

This sounds negative but actually creates freedom. Most decisions are made by the default—I don’t have to deliberate on everything. The cognitive load drops dramatically. And the things that do pass the filter get real attention instead of fragmented attention.

The filter also becomes self-tuning. If I’m saying yes too often, I feel the consequences (fragmentation, stress, incompletion). If I’m saying no too often, I feel that too (missing opportunities, stagnation). The feedback loop helps calibrate.

The Exceptions

Some categories bypass the standard filter:

Relationships. I say yes to people I want to maintain relationships with, even when the specific ask doesn’t pass the other filters. The relationship is worth the cost.

Learning. I occasionally say yes to things primarily for learning value, even when the direct output doesn’t matter. But this is deliberate and limited—I’m not optimizing for learning at the expense of completion.

Obligations. Some commitments were made in the past and need to be honored even when current-me wouldn’t make them. Reliability matters more than efficiency.

The Outcome

Constraints clarify. Say no more often.

The interesting work won’t run out. There will always be more opportunities than time to pursue them. The question isn’t how to do more—it’s how to do the right things well.

Every yes is a no to everything else. Make that trade deliberately, not accidentally.