Deep work, the version that fits in a 2026 job.
Cal Newport's book sold 1 million copies. Most readers never managed an hour of uninterrupted focus. Here's the version of deep work that actually fits the modern job.
"Just block four hours, turn off Slack, and do deep work." Sure. Have you tried that at a real company in 2026?
The 4-hour focus block is a fantasy for most people. Not because they lack discipline. Because real jobs have real interruptions, real responsibilities, and real boundaries on how much you can disappear. The deep-work canon was written for a world where the writer was a tenured professor with control over his own time.
Most of us have less control. We can still do deep work. We just have to be realistic about what it looks like.
The 90-minute version
The unit that works in real jobs isn't 4 hours. It's 90 minutes.
90 minutes is short enough that you can usually find it. Most days. Most calendars. Even Mondays.
90 minutes is long enough to do real work. Past about 20 minutes, your brain is engaged. By minute 60, you're in actual flow. The work between minute 60 and minute 90 is where the magic happens — the connections, the breakthroughs, the "oh I see it now" moments.
Two 90-minute blocks a week, every week, will outperform every "I'll do 4 hours on Saturday" promise you'll ever make.
Find one 90-minute block this week.
The protected slot
A 90-minute block is only deep work if nothing else gets in. The protection is the hard part.
Practical protections:
- Calendar block, labeled specifically. Not "Focus" — that's vague enough that people book over it. "DEEP WORK: writing Q3 strategy doc" — specific, dated, hard to override.
- Notifications fully off, not just sound. Sound off + visual notifications on is not enough. The visual flash is the disruption.
- Phone face-down in another room. Not "phone on do not disturb". The physical separation is the trick.
- One window open. One. The document or codebase you're working on. Nothing else. Slack closed. Email closed. Browser tabs closed.
Each of these by itself helps a little. Together they're transformative. The reason most "I tried deep work and it didn't work" attempts failed is that one of the four was skipped — usually the phone.
When in the day
Your prime cognitive hours are biological, not aspirational. About 70% of people peak in the morning. About 20% peak in the late afternoon. The other 10% are evening people.
Don't fight your biology. If you're a morning person, your deep work belongs at 9am. If you're an evening person, your deep work belongs at 5pm. The "wake up at 5am to do deep work" advice is great for morning people and terrible for everyone else.
The test: when you're working at peak, do you lose track of time? If yes, you're in a flow window. Schedule there.
What to do during the 90 minutes
The block needs a specific output. Not "work on the project". An output:
- Write the first draft of the strategy doc
- Solve the bug in the auth flow
- Outline the talk for next month
- Get to a working prototype of the new feature
Vague inputs ("work on X") produce vague outputs. Specific inputs ("ship the first draft of Y") produce specific outputs. The minute before the block starts, write down what done looks like. That sentence is more important than the 90 minutes.
What deep work doesn't have to be
A few myths to release.
It doesn't have to be coding. Reading, writing, thinking, design, customer research — all qualify. The criterion is whether the work requires sustained attention to produce real output.
It doesn't have to be alone. Pair-programming sessions, two-person design reviews, one-on-one strategy sessions — all can be deep work if both people show up undistracted.
It doesn't have to feel productive in the moment. Sometimes the 90 minutes feels frustrating and you "didn't make progress". That's the bottom of the U-curve. The next session usually unlocks.
The honest expectation
In a typical 2026 job, you'll get 3-5 deep work blocks per week, on a good week. That's 4.5 to 7.5 hours of actual deep work. Total.
Sounds like a small number. It isn't. Most of your peers are getting zero. The differential effect of 5 hours of deep work per week, over a year, is staggering — measured in shipped projects, books read, ideas developed, careers advanced.
The math is on your side. The discipline is the price.