Focus in the Age of Interruption

Count your interruptions today and the number will look harmless: a Slack ping here, a quick question there, a notification that took four seconds to dismiss. The ledger seems to say you lost a few minutes. The ledger is lying.

Gloria Mark, a professor at the University of California, Irvine, has spent two decades shadowing knowledge workers with stopwatches and screen logs, and her findings changed how researchers think about the modern workday. Her studies found that once interrupted from a piece of work, people took on average about 23 minutes to return to it, and often came back by way of two or three other tasks first. Her team also clocked how long people now stay on any one screen before switching: down from a couple of minutes in the early 2000s to under a minute today. The four-second ping is real; so is the 23-minute crater it leaves.

Worse, Mark found that interrupted work does not just take longer. People compensate for lost time by working faster, and pay for that speed with measurably higher stress, frustration, and sense of effort. The cost of interruption is not only where your time went. It is what the constant reassembly does to the person doing the reassembling.

The most uncomfortable of Mark's findings is who does the interrupting. Logging thousands of work hours, her studies found that roughly half of all interruptions were self-inflicted: nobody pinged, nothing rang, the worker simply broke away to check something. Constant external interruption, it turns out, trains an internal habit, until the mind starts interrupting itself in the silence. This matters for the fix: silencing your phone addresses half the problem, and the other half walks into the room with you. It is why the protocol at the end of this guide logs inner urges alongside outer pings, and why the log's usual lesson surprises people: the enemy has inside help.

There is a second cost, subtler and in some ways more important, and it has a name: attention residue. The term comes from Sophie Leroy, a business school professor at the University of Washington, who ran a series of experiments on task switching. When people moved from Task A to Task B, a residue of their attention stayed glued to Task A, especially when A was unfinished, and their performance on B dropped accordingly. You can be at the new meeting while a stubborn fraction of your mind is still drafting the half-written email.

Residue explains why a day of constant switching leaves you exhausted yet strangely unproductive: you never gave any single task a clean slice of your mind. It also explains why multitasking feels efficient and measures terribly. What feels like doing two things at once is really rapid switching, and every switch smears residue in both directions. Leroy's practical insight is that the residue shrinks when a task is closed out deliberately, even partially: writing down exactly where you stopped and what comes next lets the mind release its grip and show up whole for what follows.

Leroy's residue has an older cousin in the literature. In the 1920s, the psychologist Bluma Zeigarnik observed that people remember interrupted and unfinished tasks far better than completed ones, a phenomenon now known as the Zeigarnik effect. Unfinished business, in other words, is sticky by design: the mind keeps a file open for everything incomplete. That stickiness is useful when it reminds you to finish things, and corrosive when forty open files are each claiming a sliver of attention. The written close-out works because it gives each file somewhere to live besides your working memory.

Leroy calls her version of that close-out a ready-to-resume plan, and it takes about a minute: before switching, write where you stopped, what the next concrete step is, and anything you are afraid of forgetting. In her studies, people who made such a plan before an interruption carried less residue into the next task and performed better on it. One minute of writing buys back a measurable slice of the next hour's mind, which may be the best exchange rate in this whole guide.

Whichever interrupter you picked, it is worth knowing that each kind calls for a different wall. Notifications are the easiest, because they are a setting: they can simply be turned off for a block, and the world reliably survives it. Colleagues respond to visible signals and honest contracts: headphones as a flag, a stated 'heads-down until eleven, come find me after,' and the crucial reciprocal courtesy of batching your own quick questions for them. Meetings yield to calendar defense, holding one recurring block as unbookable before the week fills. And your own mind, the interrupter nobody can mute, responds best to the parking note: when the urge to check something surfaces mid-block, two words on paper honors the thought without following it.

The point of matching wall to interrupter is efficiency. People who try to fix focus with willpower alone are fighting all four fronts at once with the weakest weapon available. An hour spent reconfiguring notifications, expectations, and calendars buys back more attention than a month of trying harder. Fix the architecture once, and willpower gets demoted to what it is good at: the occasional gust, not the daily weather.

If switching is the disease, the cure has been named too. The computer scientist Cal Newport calls it deep work: professional effort performed in a state of distraction-free concentration that pushes your abilities to their limit. His argument is economic as much as psychological. Shallow work, the email-and-meetings layer, is easy to replicate and easy to automate; the ability to concentrate intensely on hard things is increasingly rare and increasingly valuable, which makes focus less a personal virtue than a professional asset.

Newport's key structural insight is that deep work does not happen in gaps. It has to be scheduled as blocks and then defended, because an open calendar slot is an invitation and everyone's inbox holds a stack of invitations. A block needs walls: a defined start and end, a single named target, and the interrupters physically or digitally removed in advance. Willpower is a terrible wall. Distance is a good one; a phone in another room outperforms a phone face-down on the desk, because the glance costs nothing and the residue from the glance is never nothing.

Notice what single-tasking has going for it once you see the research whole: it is not a productivity aesthetic, it is simply arithmetic. One task, held without switching, incurs no 23-minute reentry fees and accumulates no residue. The fortress below is that arithmetic turned into a routine.

Newport's other practical export is ritual. The people he studied who sustained deep work over years did not renegotiate the terms every day; they built repeatable frames: the same desk, the same starting hour, the same first action, coffee poured, note written, door shut. Rituals lower the startup cost of concentration because the sequence itself becomes the cue, the same conditioning that makes a bedtime routine work. He is equally insistent on the other bookend, a shutdown ritual that formally ends the workday: review the open loops, write tomorrow's first move, and give yourself explicit permission to stop. Focus, in his telling, is bounded on both sides or it is not sustainable at all.

One last piece, and it is the one driven people skip: rest is not the opposite of focus, it is a component of it. Attention behaves like a muscle under load, and Newport is blunt on this point: the people who produce the most deep work are rarely the ones who work the most hours, because the capacity for intense concentration tops out at a few hours a day even for the well-practiced. What protects those hours is genuine recovery around them: real breaks, an actual end to the workday, evenings where the unfinished tasks have been written down and released, Leroy-style, so they stop running background processes all night.

So build the fortress, but build the drawbridge too. One protected 90-minute block tomorrow, one honest interruption log, one real break in the middle, and a written note at the end telling your mind exactly where work stops and resumes. In an economy of scattered attention, the ability to be entirely in one place, doing one thing, is quietly becoming a superpower. It is also, not coincidentally, what a good day feels like from the inside.

And a reframe for the meeting-heavy and inbox-bound, who cannot fortress an entire day: you do not need to. The research points at a budget, not a lifestyle. Two or three protected deep blocks a week already puts you ahead of most of the modern workplace, and shallow work packs tighter when it is batched: email in two or three scheduled passes rather than a drip, quick questions collected for one corridor conversation. Guard the blocks, batch the rest, and let the interruption log tell you, week by week, where the walls need repointing. Focus was never about heroic effort; it was always about architecture.