Interstitial means "in the gap". The method, named by Tony Stubblebine around 2017, is exactly that: every time you finish one thing and before you start the next, you write two or three lines with a timestamp on them.
It looks like this:
09:12 Finished the pricing deck. Slide 4 is still weak, the comparison
does not land. Next: reply to the support backlog, 30 min.
09:47 Backlog cleared, 11 tickets. Three were the same bug.
Next: write that bug up properly. Slightly dreading it because
I do not know the cause yet.
10:31 Wrote it up. The cause was the timezone conversion, obvious in
hindsight. Next: back to slide 4.
That is the whole practice. No structure, no prompts, no app requirement. A running log of transitions, written in the transitions.
What it is actually treating
Not motivation, and not memory. It treats the specific cost of switching.
Sophie Leroy's work on attention residue found that when people move from one task to another, part of their attention stays behind on the unfinished first task, and performance on the second suffers as a result. The residue is worse when the first task was left incomplete or ambiguous.
Interstitial journaling attacks that directly. Writing "finished the deck, slide 4 is weak, the comparison does not land" does two things in one sentence: it marks the first task as concluded for now, and it parks the unresolved part somewhere outside your head so it stops running in the background. That is the same mechanism as the Zeigarnik effect, used deliberately.
The "next:" clause does the other half. Starting a task from a written intention is cheaper than starting from a blank, because the deciding already happened and deciding is where the drift begins. Without it you re-enter the day through your inbox, which is a decision made by other people.
Why it beats a to-do list at this particular job
Stubblebine framed it as a to-do list replacement, which is the part people push back on, usually for good reason. Where it genuinely wins:
- A to-do list is written by an optimist and executed by a tired person. The morning list assumes a day that does not happen. The journal is written by the person who is actually in it, with current information.
- It records the why, not just the what. A checked box tells you a task ended. A journal line tells you it ended badly, and why, which is the only version that improves anything.
- It captures the interruptions. Your list has no row for the twenty minutes that vanished into a Slack thread. The log does, because the timestamps make the gaps visible whether you like it or not.
- It surfaces resistance while it is still cheap. Stubblebine's most useful prompt is asking whether you feel resistance to the next task and why. Naming dread in one sentence is often enough to dissolve it, and when it is not, you at least learn which category of work you avoid.
Where the list still wins: anything with a deadline or a dependency. The journal is a record, not a commitment device. In practice most people who stick with this end up running both, with the journal handling the day and a real system handling the week. See note taking workflow.
Running it without it dying
The failure mode is predictable. It becomes a diary, then it becomes a chore, then it stops.
Keep entries under three lines. Two facts and a next action. If you are writing paragraphs you have started journaling in the other sense, which is a fine practice with a different purpose. See note taking versus journaling and how to journal.
Always timestamp. The times are what make the log diagnostic. Reading a week back and finding that every day has an unexplained ninety-minute crater at 14:00 is information you will never get from memory.
Always end with "next:". This is the load-bearing part. An entry without a next action is a status update, and status updates do not reduce switching cost.
One file, appended forever, or one per day. Both work. A single daily note is the natural home, and this pairs so cleanly with a daily notes habit that most people who do one end up doing the other.
Do not process it. The log is a working surface, not an archive. If a line in it turns out to matter, promote it into a real note. Everything else can rot, and should. Treating the log as material to be filed later is how you acquire a backlog and lose the habit; the distinction is the same one made in the note inbox.
The second use nobody plans for
After a few weeks the log becomes evidence, and it is unflattering in a useful way.
You find out how long things actually take, as opposed to how long you estimate they take. You find out that the task you have rescheduled eleven times is not going to happen and should be deleted. You find out which hours produce real work and which ones produce activity.
That is enough material to make the weekly review worth doing, and it is why people who keep an interstitial log usually have a better weekly review than people who do not. Most reviews fail because there is nothing concrete to review. Here there is a timestamped record of the week as it happened rather than as you remember it, and the gap between the two is the whole point.
Is it for you
Probably yes if your days are fragmented, if you lose time to switching, or if you finish a week unable to account for it.
Probably no if you already work in long uninterrupted blocks, in which case the interruption to write the entry costs more than the residue it prevents. If you spend your day in deep work sessions, journal at the boundaries only and leave the middle alone.
More in the capture habit, context switching, and the wider personal knowledge management section.