Your brain quits every 90 minutes

Set a timer this morning to test the 90-minute rule on myself before writing a single word of this issue. Ninety-three minutes later I had a complete first draft. Either the science works, or I've been successfully manipulated by my own newsletter topic. Possibly both.

In the 1950s, sleep researcher Nathaniel Kleitman — the man who discovered REM sleep — noticed something strange. The 90-minute cycles he was tracking through the night didn't stop when people woke up. According to a detailed breakdown of Kleitman's research, our brains cycle through roughly 90 to 120 minutes of high alertness followed by 15 to 20 minutes of reduced cognitive capacity, all day long. He called it the Basic Rest-Activity Cycle. Most people have been ignoring it ever since.

The practical consequence is blunt. Push past that 90-minute window without a break and your brain doesn't just slow down — it starts compensating. You feel the fog, reach for coffee, grind through another 45 minutes of mediocre output, and then wonder why you feel wrecked by 4pm. That afternoon slump isn't a willpower failure. It's biology presenting its invoice.

Remote workers are uniquely positioned to act on this. ActivTrak's State of the Workplace 2025 report found that remote workers already achieve 22.75 hours of deep focus per week, versus 18.6 hours for those in-office. That's more than 62 extra hours of focused work per year. The gap isn't because remote workers are more disciplined. It's because the office is an interruption machine, and home — structured deliberately — isn't.

The workers pulling ahead aren't grinding longer. They're batching their hardest thinking into protected 90-minute blocks, then genuinely stopping. Two to three of those blocks per day, separated by real breaks (not inbox checks), produces more usable output than six hours of fragmented attention. The research keeps landing on the same number. It's worth paying attention to why.

That's how many times the average worker is interrupted every single workday, according to Microsoft's 2025 Work Trend Index — one ping every two minutes during core hours. The cruelest part: half of all those meetings land between 9–11am and 1–3pm, the exact windows when most people's ultradian rhythms are at peak. We're scheduling collaboration directly on top of our best thinking time, then acting surprised when nothing substantial gets done.

A 1995 paper in Biological Psychology (Neubauer et al.) ran the statistical tests and found no significant 90-minute periodicity in cognitive performance. Zero. Which should probably get more airtime in the productivity space, where "ultradian rhythm" has become something close to gospel.

But here's what that paper doesn't disprove: working in defined, protected blocks — whatever their length — reliably beats fragmented attention. The 90-minute figure may be approximate rather than precise. Individual cycles probably range from 75 to 120 minutes, not a fixed 90. What the research consistently supports isn't the specific number. It's the principle of sustained, uninterrupted effort followed by genuine recovery.

The productivity industry loves a clean mechanism. "Your brain resets every 90 minutes" is a better marketing line than "structure your work into meaningful blocks and actually stop between them." Both point toward the same behaviour. Only one of them is honest about the uncertainty.

The problem with most focus tools is they time your work without stopping you from ignoring them. These four actually create friction.

• Freedom — Cross-device blocking that works on your phone too. If your blocker only covers desktop, you'll just grab your phone. Freedom doesn't leave that gap.

• Reclaim.ai — AI scheduling tool that automatically carves 90-minute focus blocks into your calendar and syncs with Slack to mark you as unavailable. Passive protection beats active willpower every time.

• Forest — Gamified focus timer that kills a virtual tree if you leave the app early. Users have collectively planted over 2 million real trees through the app's reforestation partnership. The social accountability feature, where leaving early kills your coworkers' trees too, is peer pressure deployed usefully.

• Focusmate — Pairs you with a live stranger for a 50-minute co-working session. The human accountability element bypasses initiation paralysis better than any solo app, particularly for people who find solo blocking hard to sustain.

Atlassian's "Team Anywhere" policy is one of the more studied examples of structured remote autonomy. The company lets employees choose where they work each day, with no mandatory in-office requirements. Great Place to Work's analysis found the policy led to measurable increases in applications and retention — but the mechanism matters more than the headline.

What Team Anywhere actually does is give workers control over when and how they structure their focus time. When you're not commuting and not subject to an open-plan office's ambient chaos, you can schedule your hardest work for your personal peak. Research by psychologist Peter Gollwitzer found that workers who pre-plan their response to specific distractions — "if a non-urgent Slack message arrives during deep work, I'll check it at 3pm" — reduce context-switching by 32%. That's an implementation intention, not a time management app. And it only works if you've already protected a block worth defending.

The companies seeing the biggest remote productivity gains aren't the ones with the fanciest tools. They're the ones that have stopped treating deep work as something that happens in the gaps between meetings.

The ActivTrak 2025 Workplace Report The dataset behind the 22.75 hours vs 18.6 hours remote focus-time gap — worth reading for the industry breakdowns. ActivTrak

Gloria Mark on attention recovery UC Irvine researcher Gloria Mark's finding that it takes 23 minutes to fully refocus after an interruption is the number that should make every meeting organiser uncomfortable. Mindful Suite

Microsoft's "Infinite Workday" report The full Work Trend Index special report, with telemetry data from Microsoft 365 users across 31 markets. Bleak, useful reading. Microsoft WorkLab

Kleitman's Basic Rest-Activity Cycle explained A clean primer on the underlying biology, without the productivity-influencer hype layered on top. Goals and Progress

Asana on building a deep work habit Practical guidance on session rituals, task pre-loading, and shutdown procedures — the mechanics that most articles skip. Asana

Re-reading Kleitman's original 1963 paper this week. He discovered REM sleep, mapped the waking rest-activity cycle, and spent decades being largely ignored by the productivity industry. Some vindications take sixty years.