During the day, the loop is small. I do the thing the system surfaced. When I finish, or when I get interrupted, I don’t re-plan, I glance at what’s surfaced now and do that. The system re-ranks as things change; I just keep pulling from the top. A good system, though, doesn’t hand you a single dictated task and demand you do exactly that one. It surfaces a small handful of the top candidates and lets you choose among them. This matters more than it looks. No scoring formula is perfect, and it doesn’t need to be, it needs to be accurate, reliably putting the right few at the top, not precise, claiming to know the single exact task you must do next. Being handed a few good options does two things at once: it keeps the decision easy (you’re choosing from five, not five hundred), and it keeps you in control, because you know things the formula doesn’t, your energy, your mood, the fact that you’re about to be near the hardware store anyway. You make the call. The system just makes sure the options in front of you are the right ones to be choosing between.

Two strategies also make this smoother, and both are just facets from Part 3 used in a slightly new way.

For big, open-ended work, the kind of task that has no natural finish line and so never feels startable, the trick from Part 2 comes into its own here. This guide is a good example. “Write productivity guide” is not a task; it’s a mountain. When I started, I didn’t even know enough about what I was doing to lay out a real plan. All I knew was that it mattered and I needed to make progress on it, but not too much, because if I’d let myself write “write the productivity guide” and started in, it would have swallowed every day for weeks and I’d have stopped doing anything else. Some projects are just too big to do all at once but too ephemeral to split up into specific tasks, and you have other things that need you. So I gave myself a time-box instead: “work on the guide for thirty minutes.” Thirty minutes is just my number; pick your own. A time-box turns an unbounded, aversive thing into something with a finish line you can actually reach, and it caps a project that would otherwise eat everything so the rest of your life still gets a turn. One small thing that helped enormously: when the thirty minutes were up, I jotted a quick note about exactly where I’d stopped and what came next, so the next block started in motion instead of from a cold, “where was I?” stall. When I’d moved the needle enough for now, I pushed it aside and let something else surface. There’s a granularity lesson buried in that story, too. Early on, when I didn’t yet know the shape of the work, I did a lot of small, concrete tasks: research this, check that, look into the other thing. Those earned their place, they were real things I’d have lost track of otherwise. But once I found the rhythm of writing and rewriting, a task like “finish Part 1” stopped being useful. I was already in motion, moving from one part to the next; a task telling me to do the thing I was already doing would have been pure ceremony. That’s the rule of thumb: a task earns its place when it captures something you’d otherwise lose or forget. Once you’re in flow on a project, tasks that just narrate what you’re already doing are noise, and you can let them go. And when the block ends, I do something that feels counterintuitive: I go back to the system and ask whether this is still the right thing, rather than just barreling on. Momentum is not the same as importance. Maybe continuing is right; maybe something more pressing surfaced while I was heads-down. Letting the system re-surface, even mid-task, is the whole philosophy of this guide applied to my own attention. The right task finds you, even when you’re already working on something.

The other pattern is a cooldown. Some days I’ve simply had enough of a particular task, I’m not blocked, I’ve just moved the needle on it enough and am done with it for now. Instead of letting it sit at the top of my list glaring at me, I push its start date out a few hours, or to tomorrow. Same start-date facet, but used for my own sustainability rather than for an external constraint: “I’m choosing to rest this until later.” It disappears, I work on something else, and it comes back when I’ve got fresh energy for it.