What deep work needs
Cal Newport's Deep Work describes it as focused, distraction-free work on something cognitively demanding: writing, coding, analysis, research. It needs long, protected stretches, because hard problems take a while to load into your head and every interruption sends you back to the start. Ninety minutes is long enough to get properly into the problem and short enough to protect from meetings.
How to use the 90 minutes
Decide the task before you press play and write it down; deciding mid-block is how the block gets lost. Keep one window open for the work and this tab for the timer. When attention dips, and it will, stay with the task in a lighter way rather than switching. One or two blocks a day is a lot of deep work. How much is realistic is covered in how many hours a day you can really focus. If 90 minutes is too long today, the 52/17 timer is a good middle step.
Newport, C. (2016). Deep Work: Rules for Focused Success in a Distracted World. Grand Central Publishing.
Pick a time and defend it
Deep work tends to fail in the calendar before it fails at the desk. Book the block as a real appointment, ideally at the same hour each day, and decline what collides with it. A set slot also removes the daily decision about when to begin.
A programmer's morning
- 8:30: write down the bug or feature and the test that will prove it done.
- 8:35–10:05: the block, chat closed, notifications off.
- 10:05–10:20: walk.
- 10:20: code review and messages, in shallow mode.
Measuring it honestly
Count hours of protected work per week in the session log, and leave out everything that was merely busy. A rising count over a month says more than how any single day felt. Kiri has no site blocking, so if you need that you would pair it with a separate blocker.
Who should not start here
On-call staff and anyone whose role depends on quick replies may get nothing but frustration from a 90-minute silence. Agree a shorter window with your team, or try the 52/17 cycle.