On Deep Work in an Open-Plan World

The kind of work I find most valuable — designing something genuinely difficult, thinking through an architectural problem, writing something that requires sustained attention — does not happen in fragments. It happens in stretches of uninterrupted time where the problem can be held in working memory and turned over until the shape of it becomes clear.

This kind of work is also the kind that modern work culture is systematically hostile to.

The interruptions are not malicious. Slack messages are sent by people who have reasonable questions. Meeting invitations come from people who need to make decisions. The open-plan office was designed by people who genuinely believed that spontaneous collaboration is good (it sometimes is). None of the individual interruptions is unreasonable.

The aggregate is a workday that is mostly fragments.

I've found that the only way to do the work that matters is to be deliberate about protecting the conditions it requires, because those conditions will not exist by default.

What I actually do:

The first 90 minutes of the day are blocked before anything else claims them. Calendar closed. Notifications off. This is not always possible — some days begin with incidents or urgent decisions. But it is the target, and when it happens it is consistently the most productive part of the day.

I work in time blocks rather than task lists for deep work. A task list says "write architecture document." A time block says "9-11am: nothing else exists except this document." The difference in focus is significant.

I am honest with myself about what kind of work requires deep focus and what doesn't. Reviewing a pull request, answering an operational question, providing a quick opinion — these can happen in the fragments. They don't need a protected block. The mistake is treating everything as equally demanding of attention when it isn't.

There is also something about the quality of rest. The ability to sustain deep focus for 2-3 hours depends on actually being rested. Chronic sleep deprivation and overcommitment produce a mental state where sustained focus is not really possible — there is always a background hum of incomplete things demanding attention.

This is partly a time management problem and partly a habit of mind. The background hum is often louder than the actual workload warrants; the anxiety about things undone can be more disruptive than the things themselves. Writing the list, making a decision about what will happen when, and closing those open loops helps more than most productivity tactics.

None of this is original. Cal Newport wrote a whole book about it. What I can add is that the people I know who produce the most interesting work — not the most output, the most interesting work — tend to have worked out some version of this for themselves. They have found the conditions under which they think best and they protect those conditions, even imperfectly.

The protection is imperfect by definition. Work is social; interruptions are sometimes the right answer; not every deep focus session produces something worth the time. But attempting it is still better than not, and the cumulative effect of protecting even 5 hours a week of genuine focus compounds into more than 5 hours of value.