How to Focus Better: Conditions, Not Willpower

You sat down at nine. It is now eleven. You did not slack off, exactly. You opened the document. You read the brief again. You answered three messages, checked a number in a spreadsheet, looked something up and ended up reading two other things, came back, reread the paragraph you had written, decided it was wrong, and deleted it. You were at your desk the entire time and you were busy the entire time, and if someone asked you what you produced, you would struggle to point at anything.

That is not laziness and it is not a discipline failure. It is what happens when you try to concentrate in conditions where concentration is not actually available. Attention is not a character trait you have either got or failed at. It is a resource with requirements, and most people work all day in a setup that quietly removes those requirements, then blame themselves for the result.

One distinction first, because these two problems get mixed together constantly and the fixes are unrelated. This article is about staying with work you have already begun. If the harder part for you is opening the file at all — if the task sits there for days and you feel a small dread every time you think about it — if the problem is starting rather than staying, that is a different problem with different causes. Everything below assumes you can start. It is about what happens in the next ninety minutes.

And one calm note before the practical part: if difficulty concentrating is persistent, is a marked change from how you used to be, or is seriously affecting your work or study, it is worth raising with a doctor, because there are causes an article cannot address. That is the whole of it. The rest of this assumes the ordinary version of the problem.

What is focus, actually?

Strip away the productivity language and it is something quite mechanical: holding a limited amount of information in mind while you do something with it.

Writing a paragraph means holding the argument you are making, the sentence you just wrote, the tone you committed to, and the point you promised to reach. Debugging means holding the state of the system, your theory, and the two things you have ruled out. Reading something difficult means holding the shape of the argument so far, or every new sentence lands on nothing.

That held material is small and fragile. It is not stored anywhere; it exists only as long as you keep it going, and it takes time to build — the first ten or fifteen minutes of any demanding task are spent loading it, which is why the start of a session feels sludgy and the middle, once you reach it, feels almost effortless.

Almost everything below follows from that. Focus problems are usually problems of assembly and collapse: either the material never gets assembled, or something knocks it over before you finish.

Why does one quick message cost so much more than a minute?

Here is the uncomfortable centre of the whole subject: there is no such thing as briefly checking a message.

The check itself is genuinely brief. Twenty seconds, forty at most. But the cost is not the check. The cost is that the material you were holding — the argument, the theory, the half-formed sentence — is not paused while you look away. It falls apart. And when you come back, you do not resume. You rebuild. That rebuild takes far longer than the interruption did, and often it does not fully happen: you get most of it back, lose the specific good idea you were circling, and carry on with a slightly worse version of your own thinking without ever noticing the substitution.

Research on interrupted work has consistently found the recovery period to be substantial — not seconds, but a meaningful stretch of minutes — and that interrupted tasks tend to be completed with more strain and more errors. Describing it precisely is less useful than describing what it means for your day: if you are interrupted every fifteen minutes, you never reach the state where difficult work happens. You spend the entire day in the assembly phase. That is exactly what a busy, exhausting, empty day feels like from the inside.

It also explains something that otherwise makes no sense: a day of eight hard hours that produced nothing, and a day of three hard hours that produced everything. The difference is rarely effort. It is how many times the material had to be rebuilt.

The interruption you cause yourself

External interruption gets all the attention, and it is only half the story. The larger half is that you interrupt yourself more often than anyone else interrupts you, and nobody talks about it because it does not feel like an interruption. It feels like a decision.

Notice what happens when the work gets slightly hard. A sentence will not come out right. The code does not do what you expected. The number does not reconcile. In that moment there is a small, specific discomfort — and your hand moves to the phone before any conscious thought occurs. No notification. Nothing to check. You unlock it, look at something, lock it, put it down, and could not tell anyone what you just looked at.

That reflex is not a moral failing. Difficulty is uncomfortable, and you have spent years learning that a small check reliably makes the discomfort go away for a moment. It works. That is why it repeats. The price of that moment’s relief is the entire structure you had built in your head.

Two things help more than trying to want it less:

Increase the distance. Not to another app or another screen — to another room, or at least across the room, out of sight and out of arm’s reach. The reflex operates below deliberation, so willpower is the wrong tool. Physical distance inserts enough friction for a conscious thought to arrive.
Catch the moment, not the habit. When you notice your hand moving, you have just been handed useful information: the work got hard right there. That is usually the exact spot where the real problem is. Staying in it for another two minutes is often worth more than the next hour.

What can you do if you cannot change the office or the messaging?

Neither is a neutral backdrop. A room full of movement and half-audible conversation keeps requesting your attention, and overheard speech is especially disruptive to language work because it competes for the same machinery. Messaging adds a second layer: where fast replies are expected, the possibility of interruption is continuous, and merely knowing you might be pinged keeps part of your attention on standby.

Most people cannot change either. What tends to be available anyway:

Take the good hours rather than asking for them. One protected stretch — early, late, or over a lunch that is quiet — usually exists. It rarely needs permission. It needs to be used for the hardest thing rather than for email.
Make your availability legible. “Heads-down until eleven, back to you straight after” prevents almost all of the guilt and most of the follow-ups. Colleagues generally do not mind waiting; they mind not knowing.
Batch replies instead of streaming them. Three deliberate passes through messages beats forty reactive glances, and nothing arrives meaningfully later.
Move the demanding work elsewhere. A different room, a different chair. Environment change is often faster than habit change, and headphones do real work as a visible signal.

The point is not to build a monastery. It is to stop expecting deep work to happen inside a stream of small demands, and to stop reading the failure as personal.

Why can you not concentrate on “work on the report”?

This is the most fixable cause of poor focus and the most consistently overlooked. A vague task cannot be concentrated on, because there is nothing specific to hold in mind.

“Work on the report” is not a task. It is a topic. When you sit down to it, part of your attention goes to deciding what you are actually doing, and that deciding never finishes, so it keeps interrupting the doing. From the inside that feels like drifting. What it is, is an unspecified task.

Compare:

“Work on the report” → “Draft the second section, about last quarter’s numbers, roughly 400 words, from the figures already in the spreadsheet.”
“Study for the exam” → “Work through questions 4 to 9 without looking at the answers.”
“Fix the bug” → “Find out which of the three functions changes the value, by printing it at each step.”

The second version of each can be held in mind whole. You know what you are doing and what done looks like, so all of your attention goes to the work rather than to defining it.

A useful test: if you cannot say what the next twenty minutes will physically consist of, you do not have a task yet — and the definition work is a separate job. Do it with a pen, away from the file, when your attention is already gone. Ten minutes of specifying tomorrow’s tasks buys you a far better tomorrow than ten more minutes of pushing at today’s.

Is multitasking real, and does single-tasking actually help?

For anything demanding, no — it is not parallel processing, it is switching. You can walk and talk at once because one of those runs without attention. You cannot write an email while following a discussion, because both need the same limited pool. What happens is rapid alternation, and every alternation carries the same rebuild cost as any other interruption, in miniature, hundreds of times.

This is why an afternoon of “keeping on top of things while working on the thing” produces so little. You did not do two jobs at once. You did one job badly, in fragments, with a tax on every fragment.

Single-tasking helps, but the version that helps is more specific than “do one thing at a time”:

One task, one window, one surface. Reference material open, everything else closed — not minimised, closed.
A scrap of paper next to you for the thoughts that arrive mid-task: call the bank, check that figure, reply to Sam. Write them down and carry on. Unwritten, they circle back every few minutes and cost you something each time.
Accept that the first stretch will feel unpleasant and slow. That is the assembly phase, not evidence that today is a bad day.

How much does being tired explain?

More than most of the techniques combined, and it deserves less space here than it deserves in your life, because there is not much to argue about.

Short sleep does not mainly make you feel sleepy. It narrows exactly the capacity this article is about: holding material in mind, resisting the pull of the more interesting thing, noticing you have drifted. On short sleep you can still do easy work for hours, which is why the deficit hides. What disappears is the hard work specifically.

If your evenings routinely end with a long stretch of scrolling in bed, that is worth looking at. What phones do to your attention in the evening is a bigger factor than most people credit, and changing it is behavioural rather than heroic: the phone charges in another room, and the last half hour before bed ends when you put it down.

The other physical basics are worth keeping in view and not worth over-thinking. Moving between working sessions helps most people restart more cleanly than sitting through. A very heavy lunch makes the following hour harder; going a long time without eating makes everything feel more effortful. Mild dehydration is unpleasant and easy to fix. A properly bright room beats a dim one in the afternoon. None of that is a protocol and none of it needs a product — treat it as removing obstacles, not gaining an edge.

Have phones destroyed your attention span?

You have probably wondered. The honest answer sits between the two loud positions, and both loud positions are wrong.

The doom version says your attention has been permanently damaged and cannot be recovered. There is no good basis for that. The dismissive version says nothing has changed and you should just try harder. That does not match anyone’s experience, including yours.

What has genuinely changed is practice. Sustained attention is a capacity that improves with use and fades without it, like most capacities. It used to get practised by default — in queues, on buses, waiting for things, being bored in a waiting room — because there was nothing else to do. Now every one of those gaps is filled. The capacity has not been destroyed; it has been under-exercised for years, and something under-exercised feels bad when you first ask it to work.

That reframe matters practically. If it is damage, the answer is grief. If it is a lapsed capacity, the answer is training, and training rules apply: start below your current limit rather than above it, go often rather than heroically, and expect the difficulty to be worst at the start. Twenty-five uninterrupted minutes, several times a day, done consistently for a few weeks, will change what you are capable of more than one ambitious three-hour attempt that collapses in twenty minutes and convinces you that you are broken.

Leaving some gaps unfilled helps too — the walk without headphones, the queue without a screen. Not as self-denial. As reps.

Do time-boxed working sessions work?

They help, they are not magic, and the standard fixed interval is not sacred.

What actually does the work in a time-boxed session is not the timer. It is that a defined block makes “just check something” feel like a violation of an agreement you have made, and it gives you a legitimate end point, which makes starting easier when the task is unappealing. The timer is a fence, not a stimulant.

Where the standard short interval falls down is on work with a long assembly phase. If it takes you twenty minutes to load a complex problem, a twenty-five minute block ends roughly when you become useful. For that kind of work, longer blocks with fewer breaks fit far better — and if you are genuinely deep in something when the timer goes, ignoring the timer is the right call, not a lapse.

Some things that make blocks work better in practice:

Name the block by its task, not by its length. “Ninety minutes on the second section,” not “ninety minutes of deep work.”
Take the break away from a screen. A break spent scrolling does not restore anything; it is the same activity with worse content.
Do not stop at a clean ending. Stopping mid-sentence or mid-step leaves an obvious handle for tomorrow, and the assembly phase is much shorter when there is something concrete to grab.
Match the block to the work. Short blocks suit admin, email and shallow tasks. Long blocks suit thinking. Using one length for everything is why the technique fails for so many people.

If you want this to become the default rather than an effort every day, the mechanics of building a habit matter more than the specific interval you choose: same time, same place, small enough to survive a bad week.

Should you work with music or in silence?

It depends on the task and on you, and anyone who tells you otherwise is generalising from their own head.

The one fairly reliable pattern is about language. Lyrics compete with work that involves words. Writing, editing, reading anything difficult, or composing a careful reply all use the same language machinery that processes the singing, and they interfere. Instrumental music, or something so familiar it has stopped carrying information, tends not to.

For work that is not language-based — spreadsheets, design, routine processing, repetitive tasks — music often helps, largely by making the tedium tolerable enough that you stay in the seat.

Total silence is not automatically best either. In a room with unpredictable noise, steady background sound is usually preferable to silence punctuated by other people’s conversations, because unpredictability is the disruptive part, not volume. Masking a noisy room with something even and boring works well. So does anything that keeps you from hearing exactly one half of a phone call, which is unusually hard to tune out.

Test it properly rather than assuming: same kind of task, one week each way, and judge by what you produced rather than by how it felt.

What should you do when your focus is gone for the day?

Some days it does not come back. You have had four hours of good work, or three bad nights, or a morning of interruptions that shredded every attempt. By three in the afternoon the material will not stay assembled no matter how sincerely you sit there.

The worst option is the most common one: staying at the desk, attempting the hard task, failing at it slowly, and paying in guilt for hours. That produces no work and no rest, and it teaches you that sitting at your desk is a place where you fail.

Better is to recognise it early and switch deliberately to low-demand work, of which there is always plenty:

Email, filing, admin, expenses, scheduling — real work that does not need much held in mind.
Tidying the workspace and closing the loops of the day.
Preparing tomorrow: writing the three specific, defined tasks you will start on. This is the highest-value thing a depleted brain can do, because task definition is easy work that makes tomorrow’s hard work much easier.
Reading something adjacent and undemanding.
Stopping, if stopping is genuinely available.

Calling it early is a skill. The distinction to learn is between this is hard and this is not going anywhere. The first is normal and should be pushed through. The second has a particular quality: you read the same line several times, you cannot remember what you decided ten minutes ago, and every small choice feels heavy. Once that is clearly the state you are in, more sitting does not help.

Is the real problem that you cannot tolerate being bored?

Quite often, yes — and this may be the least comfortable and most useful idea here.

Hard work has a shape. There is an interesting beginning, where the problem is fresh. There is a satisfying end, where it comes together. And between them there is a long, flat middle where the work is neither novel nor nearly finished, where you are grinding through the part that is simply effortful. That middle is where things get done, and that middle is boring and mildly unpleasant.

Most lost focus happens right there. Not because you lack ability, but because the middle feels bad and relief is one gesture away. If you can end a mildly unpleasant feeling instantly, at any moment, for free — you will. Anyone will.

So the real capacity is not attention in the abstract. It is tolerance for the unpleasant middle. And that is trainable in the same unglamorous way as anything else:

Name it when it arrives. “This is the boring middle, this is what it always feels like here.” Naming it turns an urge into an observation, which is far easier to sit with.
Give it a number. Ten more minutes, then you may decide. The urge nearly always passes inside that window, and afterwards you are usually back in the work.
Take away the escape route. Tolerance is not really trainable while instant relief is within reach. This is the practical case for the phone being in another room — not virtue, mechanics.
Notice what you get for it. The work you are proudest of almost certainly came from a stretch that felt tedious while you were in it. That was not a sign you were doing it wrong.

Nothing in this article asks you to want it more. Wanting it more was never the missing piece. What is missing is usually a defined task, a stretch of unbroken time, a phone that is out of reach, enough sleep, and a willingness to stay in the dull middle for ten minutes longer than feels natural. Those are conditions, and conditions can be changed. Character does not have to come into it.

Protecting your attention eventually means protecting your time from other people, which is a matter of making trade-offs visible.

Concentration is only half of it. Holding on to what you read afterwards is a separate skill with its own method.

If concentration has been poor for a long stretch and stress is the likelier cause, the American Psychological Association has useful material on it.

If the difficulty has lasted and feels heavier than distraction, the NHS mental health pages are the right place to start.

Common questions

How long should I be able to focus for?

There is no correct number, and the ones that circulate are mostly invented. What is reasonably consistent is that most people can do genuinely demanding cognitive work for a few hours a day, not eight, and that it comes in stretches rather than one continuous run. If you get two or three good blocks of an hour or so, you are doing well by any realistic standard. Judge yourself against output, not against hours occupied.

Why can I focus for hours on things I enjoy but not on work?

Because enjoyable activities usually supply their own structure — clear goals, immediate feedback, constant small rewards, and no ambiguity about what to do next. Most work supplies none of that. This is not evidence that your attention is fine and your motivation is broken. It is evidence that the work needs the structure added deliberately: a specific defined task, a bounded block of time, and a visible end point.

Does turning off notifications solve it?

It solves half. It removes external interruption, which is real and worth removing. It does nothing about self-interruption, which is more frequent for most people. A silent phone on the desk still gets picked up dozens of times a day. Distance works better than silence.

Is it better to work in long stretches or short bursts?

It depends on how long the work takes to load. Work with a long assembly phase — writing, analysis, complex problems — is badly served by short bursts, because the block ends around the time you become effective. Work that needs little loading — admin, email, small discrete tasks — fits short bursts well. Sort your tasks by that question rather than applying one interval to everything.

I have tried everything and nothing works. What now?

Check the foundations before the techniques, because techniques cannot compensate for missing conditions: are you sleeping enough, are your tasks actually defined, and do you have any stretch of time that is genuinely uninterrupted. Most “nothing works” situations turn out to be one of those three, and no method fixes them from the outside. And as noted at the start, if the difficulty is persistent, is clearly different from how you used to be, or is seriously affecting your work or study, that is worth a conversation with a doctor rather than another article.

How to Focus Better When Trying Harder Has Stopped Working