What to Write in a Journal, and What You Get Back

The writing itself, which puts a circling thought into one order and gets a feeling out of you. And the going back, which is where a record pays. Most advice covers the first and quietly assumes the second.

The first half is immediate and not in doubt. A thought going round your head has no order. One sentence forces an order on it. A resentment written out often ends there, rather than following you around for three days. And the reflecting is the act itself, not a document you produce at the end of it. This is what most people are there for. It works on the night.

The second half is what people who have kept journals for twenty years talk about. Checking what happened against a memory that has quietly rewritten itself. Catching a thought that would be gone in a week. Seeing a pattern invisible from inside a single day. Noticing, years later, what was in front of you at the time.

The consequence is practical. The two halves do not need the same entries. A page of pure feeling collects the first payment in full and almost none of the second, which is my grandpa's regret stated exactly. That does not make it a wasted page. It means the second payment has to be bought deliberately, and it costs about one extra line.

There is a third thing, and it needs both halves at once. Compare what actually happened against what you remember feeling about it. The gap is usually more interesting than either side alone. Almost nobody gets to run that comparison, because it needs an objective account of the day sitting next to your version of it, and by the evening the objective account is gone.

Four things end journals and none of them is a discipline problem. Each is a supply problem: at the moment you sit down, the material you need has already gone out of reach.

The first is the one that produces the blank stare. You open to a prompt, something like what mattered today, and nothing arrives. Not because you have nothing to say. By ten at night the detail is out of reach. So is the harder one: what you actually felt at the time, rather than what you feel about it now. Four or five words into your phone during the day recovers a lot of this. Pastok recovers the rest by holding the day for you, as a timeline of what happened, the highlights, and the transcript. At ten o'clock the sentence somebody actually said is still there to write from. That is also what makes the comparison above possible, because now you have a record to hold your version against.

The second is that the thinking and the writing happen in different places. "When I think about journalling, I have amazing insights, but as soon as I get my pen, my mind just goes blank." You were composing paragraphs on the walk home. The fix is to capture where the thinking is. For most people that means talking rather than writing, and Pastok starts a session from the watch face with one tap. You can also mark something mid-session: say "Pastok" and what you want kept, and it lands in Notepad in that day's report.

The third produced the regret at the top of this page. Describing what happened is work. On a bad evening you have neither the energy for it nor any wish to do it, because what you wanted was to vent. So you vent, and you are back at my grandpa's choice: outlet or description, not both. With the description already done for you, that choice stops existing. Write purely the outlet if that is what you need. The event layer is underneath it either way.

The fourth is a loop, and the reason people give most often. The less you journal, the less you can think of to write about. The more you journal, the easier it comes. So a busy fortnight makes the fifteenth entry harder than your first. Add the evenings when ten minutes alone is not realistic. Both shrink when the day arrives already assembled, because then you are editing rather than producing, and the supply does not depend on whether you wrote yesterday.

Two things decide whether Pastok is an option at all: it needs an Apple Watch and an iPhone, and it is free for the first fourteen days and paid after that. On the privacy question that has ended a lot of paper journals, this is an authenticated account rather than a notebook on a shelf, encrypted in transit and at rest, and never used to train AI models. Raw audio is deleted after processing, within seven days by default, so what you keep is the text.

For the second payment, write what you could not rebuild later: the sentence that landed, a dated prediction, the thing you nearly did. Feelings are worth keeping, attached to those rather than instead of them.

None of this is a rule. There is no version of journaling you are doing wrong, so write however you like. But if you want the entry to still be worth something in five years, these are the lines that carry.

• The event at sentence level. Not "difficult conversation with M", which tells you nothing you did not already know. The sentence that actually landed, and what you said back.
• What you expect to happen. This is the highest-return line there is and almost nobody writes it. You think the hire will not work out; you think they will call by Friday. Rereading then stops being nostalgia and becomes finding out whether your judgement is any good.
• What you nearly did and did not do. The email you drafted and deleted. Nothing happened to anchor it, so it goes faster than anything else and is worth the most later.
• What you did not understand yet. The open question you were carrying that week. Read back, it shows what was already in front of you and what you genuinely could not see.
• Feelings, sitting under the event rather than in place of it. Anchored to something described, a feeling still means something in twenty years.

One test tells you whether there is enough detail. Read the line back and ask whether someone who was not there could picture it. If not, add exactly one specific: a name, a number, a place, or something said out loud. Two or three lines like that is a complete entry. Nothing here needs twenty minutes.

If you do have a record of the day, read it before you write anything. Then put down what you remember feeling. Where the record and your memory disagree is very often the entry. You had written off a whole Tuesday as wasted, and the timeline shows the one conversation in it that actually mattered.

Two more only work once there is some history behind you. Search by meaning rather than by the exact word when you want the thought you had in March, since the word you used then is rarely the word you would use now. And write a skill that reads back over a window you choose, a month or a quarter. Run it every week and have what it finds land as a section of your daily report. A file you must remember to open is a different thing from a section that is simply there.