What should I write in a journal before bed to fall asleep faster?

A specific to-do list. In Scullin and colleagues' 2018 polysomnography study of 57 healthy adults, the group that spent five minutes writing tasks for the coming days fell asleep in about 16 minutes, against 25 for the group that wrote about completed tasks, and the more specific the items, the faster the onset. Follow it with a closing sentence, one concrete task for tomorrow, and one good thing, ten minutes in all.

What should I not write in a journal at night?

Grievances, unsent letters, trauma, and problems with no answer available tonight. The bedtime evidence is for a specific list. Deep emotional writing is a different tool: in Mosher and colleagues' 2012 trial of 62 women with metastatic breast cancer, home sessions of emotional writing produced no overall benefit and increased sleep disturbance in the longer-diagnosed subgroup. Keep that kind of writing for daylight, when there is time to finish it.

How long should bedtime journaling take?

Ten minutes, timed: three for the dump, two for the closing sentence, two for tomorrow's one task, three for one good thing. The Scullin study used five minutes of writing. The timer matters because an untimed session drifts into rumination, and the good-thing step drifts into a gratitude list, which is a second session. When the timer ends, the pen goes down whether or not the page feels finished.

Does keeping a sleep diary help you sleep?

On its own, no. Tollånes, Nielsen and Parsons (2025) randomised 190 healthy adults to a week of either a sleep diary or a physical activity diary and found no change in perceived sleep quality from the diary itself. A sleep diary is a measurement tool used inside CBT-I. The two-week log on this page is diagnostic in the same way: the four-step writing sequence is the intervention, and the log shows whether it is working.

Bedtime Journaling for Better Sleep: A Ten-Minute Wind-Down

Q: How long does bedtime journaling take to work?

The to-do list effect in the Scullin study appeared on the first night, since it measured a single night in a sleep laboratory. For a habit, Arigo and Smyth (2012) found less reported sleep difficulty eight weeks after three 15-minute writing sessions, though that was daytime writing about body concerns. Give the sequence two weeks with the log, and tighten the dump to specific items if night three is still slow.

In the ten minutes before bed, write four things in this order: everything on your mind as a list of specific items (three minutes), one sentence that closes the day (two minutes), tomorrow's single most important task as a concrete action (two minutes), and one good or neutral moment from the day (three minutes). Do not write about a grievance, an unsent letter, a trauma, or a problem with no answer tonight. The first step carries the evidence: in a polysomnography study of 57 adults, five minutes of writing a specific to-do list cut sleep onset from about 25 minutes to 16 (Scullin and colleagues, 2018).

Quick Answer: Dump, close, one task, one good thing: ten minutes, each step timed. Write the dump as specific items. Write the closing sentence even when it feels untrue. Write one task for tomorrow as an action you could start. Write one thing that was fine. Then stop.

Key Evidence: Scullin, Krueger, Ballard, Pruett and Bliwise (2018), Journal of Experimental Psychology: General: 57 adults, five minutes of writing before lights out, sleep onset 15.8 minutes for the to-do list group against 25.1 for the completed-list group, Cohen's d = 0.63.

Context: The content decides the result. Specific future tasks helped; emotional depth is a different kind of writing, and one trial of it found worse sleep in a subgroup.

Key Takeaways

Timer on every step: 3, 2, 2, 3 minutes. Ten total, then the pen goes down.
"Email Dr. Reyes about Thursday" beats "work stuff"; specificity was the variable that predicted faster sleep onset in the Scullin study.
If the dump is all worries, convert each one into a task with a time or a parked item with a date.
Cap the good thing at one; a gratitude list at 11pm is a second session.
The two-week log is diagnostic. Keeping a sleep diary by itself changed nothing in a 2025 trial of 190 adults.

The four steps

Step	Minutes	Write this	Example line	Do not
1. Dump	3	Every open item, one per line, as specific as you can make it	"Book the car in before Friday"	Write "everything" or "life"; a vague item stays in working memory
2. Close	2	One sentence that ends the day	"Today is done. I did what I could with it."	Grade the day, list what went wrong, or relitigate a conversation
3. One task	2	Tomorrow's most important task as an action with a start	"9am: draft the first paragraph of the proposal"	Write a list of ten; one item is the offload, ten is a second dump
4. One good thing	3	One moment that was good, or simply fine	"The bus was on time and the seat by the window was free"	Manufacture gratitude or keep going past one

An illustrative night, invented: the dump has five items (library books, reply to Sam about Saturday, the dentist reminder, the thing Mark said in the meeting, the unpaid invoice). The closing sentence: "Today is done and the invoice can wait until morning." The task: "8.30am: send the invoice with the corrected date." The good thing: "the walk back from the station in the last of the light." One dump item was a feeling dressed as a task; the section after next handles it.

Why the order matters

The dump goes first because open tasks are what the pre-sleep mind rehearses, and Scullin's team found that writing them as specific future items shortened sleep onset while writing about the past did not. The closing sentence is a boundary. The one task turns the largest tomorrow item into a plan with a start time. In my own executive years, meeting notes, board preparation and task lists never went into the journal; they lived in minutes and lists, and the journal was for thinking.

If the dump is all worries

Some nights step 1 produces only anxieties. Convert each line. A worry with an action becomes a task with a time: "call the bank at 9." A worry with no action available tonight becomes a parked item with a date: "nothing to do until Thursday: parked." The conversion is the offloading. Specificity predicted faster sleep onset in the Scullin study, and a parked item with a date is the most specific form a worry can take.

What not to write at night

Deep emotional writing is a different tool. In a randomised trial of 62 women with metastatic breast cancer (Mosher and colleagues, 2012), four 20-minute home sessions of writing about cancer-related emotions over three weeks produced no overall benefit, and the women diagnosed longest showed increased sleep disturbance. The trial did not specify a time of day, so the rule here is the site's own: keep emotional processing, grievances, unsent letters and open-ended problem solving for daylight, when there is time to finish them.

When this does not work

Still awake after 30 minutes on night three. The dump was vague. Tighten every line to an item you could start.
The closing sentence feels false. Write "Today is done" anyway. It is a boundary, and boundaries do not need to be verdicts.
One good thing keeps turning into a list. Cap it at one. The timer is the rule.
Sleep onset improved but mood on waking is unchanged after two weeks. Keep going. Sleep onset was the target; the mood field is diagnostic.

The two-week log

Copy this block into the back of the notebook and fill in one line a night, after the sequence.

Date: ____
Sequence started at: ____
Estimated time to fall asleep (noted next morning): ____ min
Mood on waking (1 to 5): ____
One-word theme of the dump: ____

At the end of week two, read the theme column against the sleep-onset column. A theme that appears five nights running is where the pre-sleep load is coming from. The sequence is the intervention and the log is its record, which is a different thing from a CBT-I sleep diary, a measurement tool used inside treatment. Tollånes, Nielsen and Parsons (2025) randomised 190 healthy adults to a week of either a sleep diary or an activity diary and found no change in perceived sleep quality from the diary itself. For the habit over weeks, Arigo and Smyth (2012) found less reported sleep difficulty eight weeks after three 15-minute writing sessions in 111 college women; that was daytime writing about body concerns, so it says nothing specific about bedtime.

Why bedtime journaling matters

Lying awake is the mind rehearsing open items because it has nowhere else to hold them. Ten timed minutes give those items a page, from the loudest material to the quietest. The strongest finding in this area is nine minutes off the time it takes to fall asleep, from five minutes of writing a specific list, and the log tells you within two weeks whether it is working for you.

Sources

Scullin, Krueger, Ballard, Pruett and Bliwise (2018), Journal of Experimental Psychology: General, "The effects of bedtime writing on difficulty falling asleep" – 57 adults, five minutes of writing, polysomnography; to-do list group sleep onset 15.8 minutes against 25.1, d = 0.63; specificity predicted faster onset; the paper's introduction also summarises Harvey and Farrell (2003)
Mosher and colleagues (2012), Psychology and Health, "A randomized controlled trial of emotionally expressive writing for women with metastatic breast cancer" – 62 women, four 20-minute home sessions over three weeks, no main effect, increased sleep disturbance in the longer-diagnosed subgroup
Arigo and Smyth (2012), Psychology and Health, "The benefits of expressive writing on sleep difficulty and appearance concerns for college women" – 111 college women, three 15-minute sessions, less sleep difficulty at eight-week follow-up
Tollånes, Nielsen and Parsons (2025), Behavioral Sleep Medicine, "Does keeping a sleep diary alter the perception of sleep quality?" – 190 healthy adults randomised to a sleep diary or an activity diary for seven days; no change in perceived sleep quality