The exercise
Three good things, without the homework
The practice has a name. It is usually called Three Good Things, and sometimes What Went Well. It is about twenty years old, it has been studied more than most things in this corner of psychology, and it is why this notebook opens with three lines rather than one or ten.
It is also why a lot of people have already tried something like this and quietly stopped. That part is worth being honest about, so this page does both halves: what the research found, and the specific instruction I took out.
Where it comes from
In 2005 Martin Seligman and three colleagues published a set of experiments in American Psychologist. One of them asked people to write down, every night for one week, three things that had gone well that day, and why each one had gone well.
That was the whole instruction. One week of it. The people who did it reported more happiness and fewer depressive symptoms than a control group, and the difference was still there when they were asked again at one month, three months and six months.
The result travelled, mostly into hospitals. A pilot study run out of Duke put 228 healthcare workers through a fifteen-day version and followed them for a year. A randomised trial published in 2023 ran a digital version with health care workers and reported the same direction of effect. It is one of the few things in this field that keeps showing up when other people try it.
What the research does not say
The 2005 participants were volunteers who had come to a website looking for something that would help. They reported on themselves. And the largest effects came from the people who kept going after the week was up, who are not a random sample of anyone.
So it is a real effect and a modest one. It is not medicine, and nothing on this page is treatment.
I will not tell you that three lines a night fixes a year like the one I had. Mine did not fix anything. It changed what I was able to remember about it, which turned out to be a different and more useful thing.
The instruction that makes people stop
The classic version asks for two things at once. Name the good thing. Then explain why it happened.
The explanation is the part that breaks. On an ordinary Tuesday, why did this go well is a pleasant question. On the day the liquidators call, it is an essay you cannot write, so you skip the night. Then you skip the week.
The why also quietly ranks the entries. A win you can explain starts to feel more legitimate than a win you cannot, and the smallest lines are exactly the ones with no explanation attached. Woke up has no why. It is still the one worth keeping.
What I took out, and what I left
There is no reason field. Three lines, blank, and that is the whole form. If explaining one helps you, write the explanation into the line itself. Nothing asks you for it.
Three is where it starts, not a quota. One line is a finished night. Twenty is fine.
There is no streak, so a missed night is not a broken protocol. You can write Tuesday on Friday, or come back after a fortnight and add one line, and nothing in the app has an opinion about the gap.
And you write after the day rather than before it. This is a record of what happened, not a plan for what should.
How to do it tonight
There is no correct hour and no correct length. The version that works is the one you actually reach.
- Pick a moment you are already still — in bed, on the train home, once the kitchen is done.
- Write the first thing that comes back to you, even if it is small enough to feel silly.
- Stop at one if one is what there is.
- Do not explain them.
- Come back whenever you come back.
Questions
- Do I have to write why it went well?
- No. The original instruction asks for it and this notebook does not have a field for it. If the explanation is the part that helps you, put it in the line. If it is the part that stops you, leave it out.
- Is three a target?
- No. Three is where the exercise got its name and where the page starts. One line is a complete entry. Nothing counts them, compares them, or tells you that you came up short.
- What if nothing good happened today?
- Then write the thing that happened anyway. Woke up. Had lunch. Got through the day. It does not have to have been good, and it does not have to make the day sound better than it was.
- Is this the same as a gratitude journal?
- Close, and not the same. Gratitude asks you to feel something about what you have. This asks you to notice what happened. When you are already at the bottom, the second one is much easier to do honestly.
- Does it still work if I skip days?
- Nobody has studied the version where you write twice a month, so I will not claim a result for it. What I can say is that a record with gaps in it is still a record, and a record you abandoned because of one gap is not.
Sources
- Seligman, Steen, Park & Peterson (2005). Positive psychology progress: empirical validation of interventions. American Psychologist, 60(5).
- Sexton & Adair (2019). Forty-five good things: a prospective pilot study of the Three Good Things intervention. BMJ Open, 9(3).
- Sexton & Adair (2023). Three Good Things digital intervention among health care workers: a randomized controlled trial. Annals of Family Medicine, 21(3).
If you already tried one
Gratitude helped. Then it became another thing you had failed to do.
The other guides
ADHD and difficult seasons
A journal that survives the gaps
What to change about the practice when you already know the gaps are coming.
Work and progress
A daily wins log for work
For the day that went well and reads as nothing by nine at night.
Private by default
Why the notebook is private
What an audience does to a record, and exactly what private means here.