You sat down for an hour. You tried three approaches, none of them worked, and at the end you deleted everything.
The instinct is to leave the log blank. Nothing happened, so there's nothing to write, and writing tried three things, all bad looks like documenting your own failure.
Leave it blank and you've made two errors, one of which will cost you months.
What actually happened in that hour
Something did, and it wasn't nothing.
You ruled out three approaches. That's real information — the search space is smaller than it was, and you now know things about the problem you didn't know at eight o'clock. Blind alleys are how most difficult work proceeds, and someone who has eliminated three options is genuinely further along than someone who has eliminated none.
You also kept contact. The material is back in your head, warm, available. Tomorrow's session starts from somewhere rather than from an archive, which is worth more than the deleted output was.
And the deletion itself was a judgement. Recognising that something doesn't work is a skill and it's the skill improving. It just doesn't leave anything behind that looks like progress, which is a fact about how progress is usually measured rather than about the work.
Why the blank is expensive
Two costs, and the second is the serious one.
The first is that you'll redo it. Without a record, four months from now you will try the second approach again, get the same way in, and hit the same wall — having lost the memory of ruling it out. This happens constantly and is invisible when it does, because the second attempt feels like a fresh idea. Tried the queue-based version, doesn't handle retries is seven words that save an evening later.
The second is what a blank does to the record as a whole. A log with entries on good days and gaps on bad ones is a highlight reel, and highlight reels lie in a specific direction: they make effort look intermittent when it was continuous. Read back in October, a month with four entries and twelve blanks looks like a month you barely worked. If eight of those blanks were hours spent getting nowhere, the record has misrepresented your own life to you, and the conclusion you'll draw — that you're inconsistent — is false and demoralising.
That's the real damage. Not the lost detail. The false story.
And the false story has consequences beyond feeling bad. Someone who believes they worked four evenings in a month makes different decisions from someone who knows they worked twelve — about whether the goal is viable, whether to drop it, whether they're the kind of person who finishes things. Those are significant decisions to be making on the basis of a record you accidentally rigged by only writing on the days that went well.
What to write instead, and when it's a lie
Short, factual, no self-assessment.
Hour on the sync problem. Tried polling, queue, and websockets. None handle the offline case.
That's fifteen words and it contains everything: the time went in, what was attempted, and the specific reason each failed. No verdict on whether the session was any good, and no apology.
Avoid the two failure modes. Don't write nothing, obviously. But also don't write a paragraph of self-criticism — wasted the whole evening, getting nowhere with this, starting to think I can't do it — because a log that grades you becomes a log you avoid on exactly the days it matters most. The entry is a description of what happened, not a review of your performance.
If you can name what you'd try next, add it. Half of unproductive sessions end with a vague sense of a direction, and that sense evaporates by morning unless it's written down.
There's a smaller version for the sessions that produced nothing because you couldn't concentrate. Hour on the report, couldn't focus, got about a paragraph. Still worth recording, and for the same reason — three of those in a week is information about your sleep, your workload or your health, and it only exists if the individual instances were written down at the time.
Some sessions really are wasted, and pretending otherwise is a kind of dishonesty this whole argument could slide into.
An hour spent rereading the same paragraph, or fiddling with formatting, or reading around the subject instead of doing it — those aren't blind alleys. They're avoidance, and recording them as worked on the chapter is a lie that will make a stalled goal look active. The counter will read zero days quiet while nothing whatever is happening.
The distinction is whether you engaged with the difficult part and it didn't work, or whether you found something adjacent to do instead. That's usually clear to the person doing it, in the moment, and much less clear a week later — which is another argument for writing it down while you know.
So write those too, and write them accurately. Hour on the chapter, mostly rearranged existing paragraphs, didn't touch the case studies. That's an uncomfortable sentence, and three of them in a fortnight is the most useful diagnosis you'll get about that goal. A log that only records honest effort will miss the exact pattern you most need to see.
Eleven entries in a bad month
Someone is six weeks into a difficult piece of consulting work. February goes badly and by the end of it she's convinced she achieved nothing.
The log says otherwise. Eleven entries. Four are ordinary progress. Three are blind alleys — a data source that turned out to be unusable, two analytical approaches that didn't survive contact with the actual numbers. Two say nothing, too tired. One says nothing, waiting on the client's finance team. One says two hours reading around the sector, didn't touch the model.
Read together that's not a wasted month. It's a month in which a bad data source was eliminated, two approaches were tested and discarded, a dependency appeared, and there was one evening of avoidance. Anyone reading that would say the work is progressing.
Her memory said the month was a write-off, because memory reports the mood and the mood was bad. The log reported the work. Without the three blind-alley entries and the two tired ones, the log would have shown four good evenings and a lot of blank space, and it would have agreed with the mood.
Which is the whole argument. The entries that felt least worth writing are the ones doing the work.
Why this matters more for hard goals
There's a pattern worth naming here, because it explains why the most important goals produce the emptiest logs.
Easy work produces output reliably. An hour of admin generates finished admin, so the log fills up. Difficult work doesn't — an hour on a genuinely hard problem produces a result maybe one time in three, and the other two are ruled-out approaches and half-formed thinking.
So a person who writes only when there's something to show will keep a full record of their easy goals and a sparse one of their hard ones. The goals that mattered most will look, in their own record, like the ones they neglected. And since a sparse record is discouraging to read, the hard goals will also be the ones that feel worst to check on, which is not a helpful thing to have arranged.
Writing the unproductive sessions corrects this. It's the difference between a record that reflects what you worked on and one that reflects what happened to come out.
The counter doesn't grade
goaldrift resets the gap when you write something. It doesn't ask whether the session went well, produced output, or moved anything forward.
That's deliberate. A counter that only cleared for successful work would be measuring results, and results arrive unevenly and often much later than the effort that caused them. What it measures instead is contact, and contact is the thing that determines whether a goal is alive.
An hour that produced nothing usable still happened. Write the line.
Related: One line is enough · Logging beats tracking · Ninety minutes is the unit
Share: