Doing more with the time you have

Write an if-then plan

The best-evidenced technique in the whole productivity literature is the implementation intention: a sentence of the form "if it's 8pm and the dishes are done, then I open the editor and write the next function." Gollwitzer and Sheeran's 2006 meta-analysis of 94 studies found a medium-to-large effect (d = 0.65) on goal attainment, over and above how motivated people already were. A 2024 update softens that number a little, but the core finding holds: naming the exact cue and the exact action works far better than intending to "make time."

Do the hardest thing first

Habbert and Schroeder (2020) ran seven experiments on task ordering. Starting with the hardest task built more confidence than warming up with easy ones, and people consistently predicted the opposite. The usual justification, that willpower is a tank that drains through the day, has collapsed (more on that below). Frog-first works anyway, just for a different reason: pre-commitment, and the efficacy boost of having the worst thing behind you.

Pick tomorrow's task today

End each session by deciding the next session's first action and writing down where you stopped. Masicampo and Baumeister (2011) showed that unfinished goals intrude on unrelated thinking, and that making a specific plan removes the interference. Leroy and Glomb (2018) tested the same move against "attention residue" and found that a one-minute ready-to-resume note measurably protected performance. Arriving with the decision already made is the single biggest recovered cost in a 60-minute block.

Estimate from history, not hope

People reliably underestimate their own completion times (Buehler, Griffin and Ross, 1994), and the fix with the best track record is reference class forecasting: look up how long the last three comparable things actually took and use that number. Your own plan for this task is the least trustworthy input you have.

With a short window, fragmentation is the main enemy. One task per session. Leroy's 2009 experiments showed that attention sticks to an interrupted task and degrades the next one, which argues against splitting an hour across two projects more strongly than it argues for any particular scheduling system.

About interruptions: the famous claim that refocusing takes 23 minutes and 15 seconds appears in none of Gloria Mark's published papers. It comes from a 2006 press interview and spread unchecked. Her actual studies found 11 to 16 minutes to return to an interrupted task, and that interrupted work often gets finished faster, at the price of measurably higher stress. The honest framing is that interruptions tax stress and quality, not that a meter runs for 23 minutes.

The distraction intervention with clean experimental support is turning notifications off. Stothart and colleagues (2015) showed that merely receiving a notification hurt performance on an attention task even when nobody touched the phone. The related claim that a phone's mere presence on the desk drains your brain is much weaker: a preregistered replication of the "brain drain" study found nothing, and a 22-study meta-analysis put the attention effect near zero. Silence the phone; the face-down ritual is mostly theater.

Cal Newport's deep work advice survives in reduced form. There is no controlled trial of "deep work" as such, but the practical core, a rhythmic same-time daily block with one defined output, matches what the attention research supports.

The 25-minute pomodoro has no scientific derivation. Francesco Cirillo tried intervals between roughly 5 and 30 minutes as a student and kept 25 because it suited him and his kitchen timer. The one randomized head-to-head trial (Smits, Wenzel and de Bruin, 2025, with 94 students in two-hour study sessions) found that strict pomodoros produced faster-rising fatigue and faster-falling motivation than self-regulated breaks, with no productivity difference at all.

Breaks themselves do help. Ariga and Lleras (2011) showed brief pauses prevent the slow decay of sustained attention, and a 2022 meta-analysis of microbreaks found reliable gains in energy and reduced fatigue, though only a small, non-significant effect on performance. Walking is a good break: Oppezzo and Schwartz (2014) found it boosted idea generation in 81% of participants, with the effect persisting after sitting back down.

For a 60-minute window the practical advice is: run it as one continuous block and take a break only when focus actually breaks. Save strict timers for tasks you are avoiding, where their real value is lowering the cost of starting. Our Pomodoro tool lets you set any interval, and the Timers page covers plain countdowns.

Sleep is the actual lever

Van Dongen and colleagues (2003) restricted people to six hours of sleep for two weeks. Cognitive deficits accumulated until they matched two full nights of no sleep at all, while subjective sleepiness plateaued early, so nobody could feel how impaired they were. Never buy your free hour by cutting sleep; the trade compounds invisibly. Take the hour from consumption instead.

Chronotype, with a caveat

The "synchrony effect" (harder thinking goes better at your circadian peak) is real but conditional. A 2025 systematic review found it in under half of adult studies, and a large preregistered study found no general effect. It matters most for strong morning and evening types and for older adults. If you are clearly a lark, put the hard hour before mid-morning; a clear owl, in the evening. The roughly half of people who are intermediate types should stop optimizing timing and just keep the slot consistent.

The well-documented part of Daniel Pink's peak-trough-recovery arc is the afternoon slump: medical errors, test scores and road crashes all worsen around 2 to 4pm. If your free time is forced into that window, give it looser work like drafting and outlining rather than debugging or analysis. And whatever your type, never spend your peak on email.

Caffeine and naps

Drake and colleagues (2013) found 400mg of caffeine taken six hours before bed still cut sleep by over an hour, and participants did not notice. With a five to six hour half-life, an afternoon coffee taxes tonight's sleep and therefore tomorrow's hour. If your slot is in the evening, a 10-minute nap beforehand is the better tool: Brooks and Lack (2006) found it gives the fastest alertness gain with essentially no grogginess, lasting up to two and a half hours.

The popular 90-minute ultradian work cycle is a scheduling convention, not a biological law. The sleep-stage cycle is real; attempts to find a matching 90-minute alertness rhythm while awake have come up empty. Its one genuine virtue is forcing a break.

Run one project at a time

Dalton and Spiller (2012) found that detailed planning helps when you have one goal and backfires across many, because the plan makes the total load salient and commitment drops. Wrosch and colleagues (2003) linked the ability to actually quit unattainable goals to better wellbeing and even healthier cortisol profiles. On one or two hours a day, two simultaneous priorities is where projects go to stall. Keep a written "not now" list so a shelved goal is a decision rather than a nagging failure.

Buy back recurring time

Whillans and colleagues (2017) found people who spent money saving time (cleaning, groceries, chores) were happier than those buying equivalent material things, yet only 2% of adults given a windfall would spend it that way. One outsourced or automated chore that returns 90 minutes a week is worth more than any scheduling app.

Batch your messages, with realistic expectations

Checking email in two or three fixed windows lowered stress in Kushlev and Dunn's 2015 study, and a 2019 field experiment found batching notifications three times a day improved mood and attention (hourly batching was worse than doing nothing). A 2022 workplace study adds the caveat: the benefits faded after about two weeks without deliberate maintenance. Keep messages entirely outside your protected block, in a short window before or after.

Habits: the honest version of "66 days"

Lally and colleagues (2010) tracked 82 people forming a single small habit. The famous 66 days is the median time to near-automaticity; the actual range ran from 18 to 254 days, and about half the participants never got there within the study. Two findings worth keeping: missing a single day did no measurable harm, so a broken chain is not a reset, and anchoring the behavior to a consistent existing cue is what drove the curve. Make the wanted action the easy one: environment staged the night before, competitors logged out, phone in another room.

Ego depletion, the idea that willpower is a fuel tank drained by use, failed two large multi-lab replications (23 labs in 2016, 36 labs in 2021, over 5,600 participants combined, effects indistinguishable from zero). The related "hungry judges" study is confounded by non-random case ordering. Stop rationing decisions; fatigue is real, the battery model is not.

Parkinson's law was satire about bureaucratic headcount in a 1955 Economist essay, not a finding about individuals. What does hold: people will impose deadlines on themselves and those deadlines reduce procrastination, but externally accountable deadlines beat self-set ones (Ariely and Wertenbroch, 2002). Tell someone when to expect the output.

The 80/20 rule is a real statistical pattern in some domains and an unsupported extrapolation in most productivity advice. Useful as a question ("which single item would make this month count?"), misleading as a ratio. And the claim that habitual multitaskers permanently damage their attention largely failed replication too; the well-supported version is simply that switching costs are real in the moment.