Time management at work

Most advice about managing time has a hole in the middle of it. It tells you how to spend your hours without ever establishing where they currently go, and quietly asks you to use your memory of last week as the baseline. Your memory of last week is not evidence. It is a summary written by someone with an interest in the result.

This is a practical guide: eight methods that hold up, how to start each one, and where each one breaks. Anything here with a number behind it says where the number came from. One well-known number is included because it is wrong, and it is the first thing worth clearing out of the way.

The statistic everyone quotes, and why it is not true

You will have read that it takes twenty-three minutes and fifteen seconds to refocus after an interruption. It is in nearly every article written on this subject, usually without a source.

It traces back to a 2006 Gallup interview with Gloria Mark, an informatics researcher at the University of California, Irvine. It was a remark in a conversation, not a published finding, and the academic papers most often cited for it do not contain the figure.

What her team did publish is more useful. In The Cost of Interrupted Work: More Speed and Stress — Mark, Gudith and Klocke, presented at the CHI conference in 2008 — people who were interrupted finished the work in less time than those who were not, with no difference in quality. They compensated by working faster. What it cost them was stress, frustration, time pressure and effort.

So the cost of being interrupted is real, and it is not twenty-three minutes of lost work. It is that you speed up to cover the gap, and that is what wears you out by Thursday. That changes what to do about it: you are protecting your capacity to keep going, not rescuing a lost half hour.

What the evidence does support: you are bad at estimating your own work

The planning fallacy was named by Daniel Kahneman and Amos Tversky in 1979: we underestimate how long our own tasks will take, even while knowing that similar tasks have overrun before. In a study by Buehler, Griffin and Ross published in the Journal of Personality and Social Psychology in 1994, psychology students were asked to predict when they would finish their thesis. Roughly a third finished inside their own estimate.

The part that matters here is that the bias survives knowing about it. You do not fix it by concentrating harder when you estimate. You fix it by having a record of how long the same kind of work took last time, and using that number instead.

Eight methods, and where each one fails

None of these is new and none of them is a system you must adopt whole. Take one. The note at the end of each is not a disclaimer; it is the situation in which that method will waste your week, and it is the part most articles leave out.

1. Collect a week of evidence before you change anything

Every method below asks you to decide what to do with your hours. None of them work if your picture of where the hours currently go is wrong, and it almost certainly is. So the first week is not a system. It is a measurement.

Record what you are doing at intervals through the day for five working days. Do not tidy it up, do not leave out the hour that went nowhere, and do not start optimising halfway through. At the end of it you will have something almost nobody has: a baseline that is not a memory.

• Pick an interval and stick to it. Fifteen minutes is fine; half an hour is enough.
• Write what you were actually doing, not what you were supposed to be doing.
• Leave gaps as gaps. An hour you cannot account for is data, not a failure.
• Mark anything you filled in later as filled in later. It matters when you read it back.

2. The Eisenhower square

Four boxes: urgent and important, important but not urgent, urgent but not important, and neither. It is old, it is simple, and it survives because the second box is where almost everything that matters lives and almost nothing ever gets scheduled.

Its real use is not sorting a to-do list. It is noticing that your week is being spent in the urgent column while the important one waits.

• Empty your head onto one list first. You cannot sort what you have not written down.
• Put every item in exactly one box. Two boxes means you have not decided.
• Give the not-urgent-but-important box a time in the diary this week, or it will not happen.
• Be honest about the fourth box. Something you keep not doing may simply not matter.

3. Block the time, then check the blocks against what happened

Time-blocking is putting the work in the diary rather than on a list. A list tells you what to do; a block tells you when, and forces you to admit how much will fit.

The half everybody skips is the second one. A block you did not keep is the most useful information the method produces, and it is only visible if you compare the plan against the record afterwards.

• Block whole pieces of work, not tasks. “Proposal”, not “write proposal intro”.
• Leave a third of the day unblocked. That third is not slack; it is the part that always happens.
• Block the admin too. If it is not in the diary it will eat the parts that are.
• On Friday, set what you planned beside what happened, and change next week's numbers rather than your intentions.

4. Pomodoro, and what it is really for

Twenty-five minutes of work, five minutes away, and a longer break after four rounds. The timer is not the point. The point is that a defined short stretch is easier to start than an undefined long one, and starting is the part that is hard.

It is at its best on work you are avoiding. It is at its worst on work that takes twenty minutes just to get back into.

• Decide what the twenty-five minutes is for before you start it.
• Leave the desk in the five. Looking at a different screen is not a break.
• If you are in flow at the bell, keep going. The timer serves you.
• Do not count the rounds as a score. A day is not better because it had more of them.

5. Batch the small things, and protect one unbroken stretch

Email, messages, approvals and small admin are not the problem individually. They are the problem because of where they land: in the middle of everything else.

Two or three fixed windows a day for the small stuff, and one stretch that nothing is allowed into, gets you the same work done in a shape that costs less.

• Two email windows a day, at times that suit the people waiting on you.
• Say when they are, so nobody has to guess whether you have seen something.
• Put the unbroken stretch where your attention is actually best, which for most people is not three in the afternoon.
• One hour is enough to start with. An hour you actually get beats a morning you defend and lose.

6. Have the sentence for no already written

Most people do not fail to say no because they lack conviction. They fail because they are asked in a corridor and have three seconds to answer, so they say yes and regret it by lunchtime.

The fix is not assertiveness. It is having said the words before, so they are available under pressure.

• “I can do that, but not this week. Would next Tuesday work?” Most requests are fine with this.
• “I can do that if something else moves. Which would you rather I put down?” Hands the trade-off back.
• “Not me, but ask — they are closer to it.” A redirection is not a refusal.
• “Let me look at the week and come back to you by five.” Buys the three seconds you did not have.

7. Do the hard thing first, unless it needs other people

Most people have two or three hours a day when they think clearly, and for most of them it is early. Spending that window on email is the single most common way a good day is wasted.

The rule is not moral. It is about matching the work that needs your best attention to the hours when you have it.

• Find your window from the record rather than assuming. It may not be when you think.
• Decide the night before what the hard thing is, so the morning does not start with a decision.
• Do not open the inbox first. Ten minutes in there sets the agenda for the next two hours.
• If the hard thing needs a decision from someone else, do the asking first and the work second.

8. Review on Friday, not in January

A time-management system that is reviewed once a year is not a system, it is a resolution. Twenty minutes on a Friday is what turns a week of data into a change worth making.

The point of the review is not to grade yourself. It is to find one thing to test next week, and to look at last week's test honestly.

• Set the plan beside the record. Where the gap is largest, ask why before you ask what to do.
• Pick one change. One. A week with four changes tells you nothing about any of them.
• Write down what you expect to happen, so next Friday you can tell whether it did.
• Keep the reviews. Six of them is a picture of your working life that nothing else will give you.

Why most of this fails anyway

The methods are fine. They have been fine for decades. What fails is the upkeep.

Every one of them asks you to do something at the exact moment you are least able to: when you are busy, late, and being asked for things. Anything that depends on you remembering, at that moment, to open a document and write something down will stop. Not because you lack discipline, but because that is what being busy means.

So the honest filter for any method on this page is not is it a good idea. It is how small is it, and what happens on the day I forget. A method that needs five minutes a day and survives being skipped on Wednesday will beat a better method that needs twenty and does not.

The same filter applies to the record itself, which is the one thing everything else depends on. If keeping it costs more than a couple of seconds at a time, it will not survive a bad week — and a bad week is precisely the week you most want the data from.

Tools, honestly

You do not need software for any of this. A paper diary and a Friday habit will beat an app you stop opening.

If you do want software, the honest lay of the land is that there are three kinds. Manual timers (Toggl, Harvest, Clockify, Tyme) are accurate when you remember to press them and blank when you do not. Automatic trackers (Timing, RescueTime, Rize, Memtime, Timemator) give you a complete record by watching which applications, documents and web pages you have open, which some people are entirely comfortable with and others — or their clients — are not. Both are covered honestly, with prices and dates, on our comparison pages.

A declaration of interest. We make a third kind: Quarter Past asks what you are working on at an interval you choose, and one tap logs the quarter-hour that has just passed. It exists because the first method on this page — a week of honest evidence — is the one everybody abandons by Wednesday. It is for the Mac, free for fourteen days, then £54 once. That is the only pitch in this article, and it is at the bottom on purpose.

If you only do one thing

Spend five working days recording where the time actually goes, without changing anything and without tidying it up. Then pick one method from the list above, run it for a fortnight, and judge it against the record rather than against how the fortnight felt.

Almost everybody skips the first part and starts at the second. That is why the advice so rarely sticks: it is being applied to a week nobody has actually looked at.