I Tracked My Actual Focused Hours for Six Weeks. The Number Was Two and a Half a Day.

For six weeks I kept a tally: every time I finished a stretch of genuinely concentrated work, I wrote down the start and end. A daily plain-text log is one way to keep those entries alongside the work they describe. Not “working” — I work far more than this — but the kind of work where I lose track of the room. Across 30 working days the average was 2 hours 34 minutes a day, and the best single day was 4 hours 10.

I had been designing my days as though the number was six or seven. Every planning system I had built assumed a majority of the day was available for demanding work, and the systems kept failing, and I kept concluding that I was undisciplined. The number says something different: I was allocating a resource I did not have. Time blocking around interruptions tackles the same mismatch between the calendar and the day.

The most immediate consequence was that I stopped scheduling more than three substantial things in a day. Not as a discipline or a minimalism thing — arithmetically, at two and a half hours, three substantial tasks is already optimistic. Every plan I had made for years had four to six on it, which meant every plan was guaranteed to end in failure, and the failure was arithmetic rather than character.

The second consequence was about placement. My focused stretches were not evenly distributed. Sixty-one percent of the total fell between 8am and noon, across the whole six weeks, and this held on days when I had deliberately tried to work late. I had known I was a morning person in a vague way; the specific version — that roughly three-fifths of my capacity lives in four hours — turned meetings before 11am from a mild annoyance into something I now decline as a matter of course, which I had never felt entitled to do before I could point at a number.

The third finding was the one I did not want. Days with any meeting before noon averaged 1 hour 48 of focus. Days with none averaged 3 hours 22. That is nearly double, and the gap is far larger than the meetings themselves, which were mostly half an hour. A meeting at ten does not cost thirty minutes; it costs the morning, because the hour before it is not usable for anything I can lose myself in, since I am waiting.

What I did with that: I now cluster every meeting into Tuesday and Thursday afternoons and I am unreasonable about it. This is standard advice that I had read many times and never acted on, and the reason I acted on it this time is that I had measured the cost of not doing it in my own numbers rather than reading someone else’s.

A caution about the method, because it does have a distorting effect. Knowing you are keeping a tally makes you slightly more inclined to push for one more stretch, which means my 2:34 is probably a small overestimate of an untracked week. I do not think this invalidates the comparisons within the data — the meeting-day gap, the morning concentration — but I would not treat the absolute number as precise.

The reason I would recommend doing this yourself rather than taking my figure is that the interesting output is not the average. It is the distribution: when your hours are, what days destroy them, and how far the reality is from what your calendar assumes. Mine was off by a factor of nearly three, and every planning failure of the previous four years fits inside that gap.