I have a bad habit I am not proud of. When the work is going badly, I keep going anyway. The screen stays on, the cursor stays blinking, and I sit there past the point where anything good is coming out, as if the act of staying put is itself the work. Some part of me treats the unbroken stretch as proof of effort. If I am still at the desk, I must still be trying. The output stopped an hour ago, but the posture continued.

So a finding that the most productive people are the ones who stop on purpose lands somewhere uncomfortable for me. It suggests my proof-of-effort instinct has the whole thing backwards.

I am not a psychologist or a productivity scientist, and this piece is me thinking out loud about one company’s data and one lab experiment, not advice. The numbers below come from particular groups of people in particular conditions, not settled rules about how your brain works.

What DeskTime actually found

In 2014, the time-tracking app DeskTime, looked at its users and isolated the top 10% most productive ones, measured by the share of their time spent on productive apps. Writing up the study, Julia Gifford described the pattern those top performers shared: roughly 52 minutes of focused work, then a 17-minute break in which they stepped fully away from the computer.

This is one study and not the final word. DeskTime repeated the exercise in 2021 and found a longer 112-minute work, 26-minute break ratio during the pandemic. So the exact numbers move.

The shape is what interests me. Because the shape is counterintuitive. The most productive tenth were not the ones grinding without pause. As DeskTime’s own write-up put it, “The employees with the highest productivity ratings, for the most part, don’t even work 8-hour days.”

Why nonstop loses

There is a thread of cognition research that fits this without being about productivity apps at all. In a study at the University of Illinois, Atsunori Ariga and Alejandro Lleras found that brief diversions during a long, repetitive task helped people sustain performance, while a control group’s focus drifted downward over time. This was a single lab task of about an hour with 84 participants, and Lleras was proposing an interpretation that challenged the older “your attention is a fuel tank that runs dry” view, so read it as a clue, not a closed case.

Still, it names something I recognise. My own honest deep-work ceiling is about three hours a day. Past that, the screen is on but the work has quietly downgraded to editing and admin, not the load-bearing producing kind. I am output-based, not hours-based, and I know this about myself. And yet I keep working past the ceiling anyway, partly out of habit, partly out of identity, partly as a hedge against the fear that three hours cannot possibly be enough. The hard stop I keep telling myself to hold is the discipline I most consistently fail. Rest, framed as part of the output, is exactly the frame I struggle to believe at 4pm with the cursor blinking.

Where the rule doesn’t fit me

I should also say the specific numbers have never worked for me. I tried the Pomodoro Technique, with its 25-minute intervals, and it never stuck. Twenty-five minutes is roughly the point where I am just getting into something, so the timer going off felt like being yanked out of the water right as I started swimming. I do better with 90-minute to two-hour blocks, which means the 52 in 52/17 is not my number either. DeskTime’s own 112 from 2021 is closer to my actual rhythm than the famous figure is.

Which makes me think the 52/17 split, repeated in productivity blog after productivity blog, is something closer to cargo cult thinking. People copy the ritual, the precise minutes, the timer, the walk, and wait for the productivity plane to land. The ritual is not the thing. The thing underneath is harder and less shareable: a focused block, then a real stop, then back.

And the harder question, the one I cannot dodge by adjusting a timer, is whether I can tell the difference between productive discomfort and the narcotic of appearing busy. One of them feels bad and produces work. The other feels virtuous and produces nothing but posture. At 4pm with the cursor blinking, I am usually not sure which one I am in. That, more than any interval, is the thing worth getting honest about.