Do Fewer Things, at a Depth That Shows
When AI makes output effectively infinite, volume stops being a moat. The real edge is doing fewer things at a visibly higher depth. Here is how to choose them.
You can now produce more in a Tuesday afternoon than you used to produce in a month. Draft the essay, generate the landing page, spin up the side project, write the emails, storyboard the video, ship the feature. The machine will keep going as long as you keep asking. Output has quietly become the cheapest thing in your workflow.
So here is the uncomfortable question nobody wants to sit with. If everyone can produce endlessly, what is any of it worth?
That is the whole essay in one line. When output is effectively free, volume stops being a moat, and the thing that separates you is depth. Not more. Better, and visibly so. The move that looks backwards right now, doing fewer things on purpose, is the only move that still compounds.
The busywork mirage
Watch what actually happens when producing gets cheap. You do not slow down and think harder. You produce more. The number of things in flight goes up, the quality bar quietly slides sideways, and you feel productive because the dashboard is full. Ten drafts, six repos, four half-built products, a content calendar that never empties.
Here is the trap. Motion feels like progress, and output is the most convincing kind of motion. You shipped something today, so it must have been a good day. But most of what you shipped was the easy eighty percent, the part the machine was always going to be good at. The hard twenty percent, the part that would have made one of those things genuinely excellent, got skipped, because you were already three tasks down the list.
I will say it flat. Most people are using their new output capacity to hide from the hard work, not to get to it faster.
“Producing more was never the bottleneck. Deciding what deserves the depth always was, and now it is the only decision that matters.
”
The mirage is that a busy pipeline is a healthy one. A pipeline full of shallow, interchangeable output is not an asset. It is noise you are paying to generate, and increasingly, noise other people can generate too, in the same afternoon, with the same tools. If a competent stranger with the same subscription could have produced your week’s work, you did not build a moat. You filled a feed.
Why depth compounds and volume doesn’t
Volume and depth behave completely differently over time, and that difference is the entire argument.
Volume is linear at best and usually worse. A hundred mediocre blog posts do not add up to one great one. They add up to a hundred mediocre blog posts, each competing for the same shrinking attention, each dragging down the average of everything with your name on it. The tenth throwaway project does not make you more known for anything. It makes you known for nothing in particular, which in a crowded field reads as invisible.
Depth compounds, and it compounds in three directions at once.
- Reputation. One genuinely excellent thing does more for how people see you than fifty adequate ones, because excellence is rare and memorable and adequacy is neither. People do not remember your output count. They remember the one piece that was clearly a level above, and that memory follows you into every room you are not in.
- Mastery. Depth is where you actually get better. Finishing the hard twenty percent, the part that resists you, is the only part that grows your skill. The easy eighty percent you could already do, or the machine could, so grinding through more of it teaches you nothing. The reps that matter are the ones at the edge of what you can pull off.
- Assets that keep paying. A deep piece of work has a long tail. The definitive guide, the tool people actually depend on, the framework others quote, these keep returning value for years while shallow output decays in days. You build one thing worth returning to and it works for you while you sleep. You build fifty things worth glancing at once and you have to keep feeding the machine forever just to stay visible.
The tax you are paying without seeing it
There is a mechanical reason doing fewer things produces better work, and it has nothing to do with willpower. It is about how attention actually behaves when you split it.
When you switch from one task to another, your attention does not switch cleanly with you. A residue of the first task stays stuck to your mind while you try to engage with the second. Researchers who study this call it attention residue, and it is a well-documented effect: the more unfinished the thing you left, and the faster you jumped, the more of it clings and the worse you perform on what you moved to. You are physically present on the new task. Part of your head is still back on the old one.
Now picture the modern shape of an ambitious person’s day. Six projects open. Notifications pulling you sideways every few minutes. Each switch drags residue from the last thing into the next. You are never fully on any single problem long enough for the deep part of your brain to engage, which is exactly the part the hard twenty percent requires. You end the day tired, having touched everything and finished nothing, and you cannot understand why the work felt so heavy for so little to show.
You do not need a precise number on the cost of a switch to feel this one. The point is directional and it is large: fragmentation is not a small tax on deep work. It is the thing that makes deep work impossible. And an environment that lets you produce endlessly is an environment that invites endless fragmentation, unless you actively refuse it.
How to choose the few things
So depth beats volume and fragmentation kills depth. The obvious response is do fewer things. The hard part is choosing which few. Cutting is easy in theory and brutal in practice, because most of your options are genuinely good. You are not choosing between good and bad. You are choosing between good and great, and saying no to a lot of good.
Three filters, in order.
Leverage: where does depth pay off the most?
Not every task rewards depth equally. Some work is capped: it can be fine or it can be excellent and almost nobody will notice the difference, because the ceiling is low. Other work is uncapped: the difference between good and great is the difference between ignored and career-defining. Spend your depth where the ceiling is high and the gap between adequate and exceptional actually changes the outcome. Let the capped work be adequate, or hand it to the machine, and stop feeling guilty about it.
Alignment: does this move the one thing?
Pick the single outcome that matters most over the next stretch, and score every commitment against it. Most of what fills your list is not wrong, it is simply unaligned: worthy work pointed in a direction you are not trying to go. Unaligned good work is the most expensive kind, because it looks productive while it quietly costs you the thing you actually wanted. If it does not serve the one outcome, it is a distraction wearing a good disguise.
Onliness: is this the thing only you can do?
Ask what you would do if the machine handled everything it could handle. What is left is your real work: the taste, the judgment, the specific angle nobody else would take. That residue is where your depth belongs, because it is the part that cannot be reproduced by a stranger with the same subscription. Choosing well here is a matter of knowing your own edge, which is why judgment is now the job. I dug into that in taste is the job.
Run something through all three and you are left with a short list, usually shorter than feels comfortable. That discomfort is the signal you did it right. If your list of priorities is long, it is not a list of priorities. It is a list of things you have not been honest about yet.
The practice: constraints, refusal, finishing
Choosing the few things is a decision. Protecting them is a daily practice, and it comes down to four habits that are simple to name and hard to hold.
Impose a hard constraint. Abundance does not self-limit, so you have to limit it. Cap the number of active projects at something uncomfortably low, two or three, and refuse to start a new one until an old one is genuinely finished. A constraint you cannot exceed does the deciding for you on the days your discipline is thin, which is most days. The point of a rule is that it works when you don’t feel like it.
Make no the default. Every yes to a new thing is a no to depth on the things you already chose, you just do not feel the second no because it is silent. Treat new commitments as guilty until proven essential. The cost of a bad yes is not the hour it takes. It is the residue it leaves on everything else, the sixth open tab in your head that makes the other five shallower.
Single-task on purpose. Do one thing at a time, long enough to cross the threshold where the hard part actually yields. That means closing the other tabs, silencing the pulls, and staying on one problem past the point where it gets uncomfortable and your instinct is to switch to something easier. The discomfort right before a breakthrough feels exactly like the discomfort that makes you quit. Learn the difference. Naming what you are feeling in that moment is its own skill, one I wrote about in naming it is a skill, and it is often what separates the people who push through from the people who bail.
Finish. A thing at ninety percent is worth close to nothing, because the last ten percent is the depth, the polish, the part that makes it visibly a level above. It is also the part your brain most wants to skip, because it is the hard, unglamorous, non-novel grind after the fun of starting has worn off. Finishing is where volume-thinking loses and depth-thinking wins. The unfinished pile is not a sign of ambition. It is a sign of too many yeses.
None of this is a productivity hack for doing more. It is the opposite, and that is the point. The instinct the moment gives you is to exploit infinite output by producing infinitely, and that instinct is exactly wrong. The people who will stand out over the next decade are not the ones who shipped the most. They are the ones who chose a few things that mattered and went deep enough that the depth was undeniable, while everyone else was drowning in their own output.
Doing fewer things is not the safe move. It is the ambitious one. It is a bet that in a world of infinite adequate, one excellent thing wins. That bet has never been more likely to pay off, and choosing what to bet on is itself a skill you can build. Which is where agency, the meta-skill comes in: none of this works if you wait for someone to hand you the short list.
Close the extra tabs. Pick the one that only you can do. Go deep enough that it shows.
