Three good things: what the gratitude research really shows.

Three good things is the exercise almost everyone has been told to try. Write down, at the end of the day, three things that went well and why. It is in school assemblies, corporate wellbeing decks and the opening chapter of most books with the word happiness on the cover. Somewhere along the way it acquired a reputation for rewiring the brain, and a counter-reputation, in some circles, as toxic positivity with a notebook.

Neither is what the studies show. The research on gratitude exercises is unusually deep for something so simple, and it tells a calmer story: a real effect, a modest one, sensitive to how and how often you do it, and not something to force on a day that was genuinely bad.

The original exercise

The version everyone cites comes from a 2005 paper by Martin Seligman, Tracy Steen, Nansook Park and Christopher Peterson in American Psychologist. They ran a placebo-controlled study over the internet with several hundred participants, each randomly assigned to one of five exercises or a control task that involved writing about early memories. The three good things group wrote, every night for one week, three things that had gone well that day and a causal explanation for each.

The week was the whole intervention. Yet when the researchers followed up at one week, one month, three months and six months, the three good things group was still happier and less depressed than the control group at six months, and the curve actually rose over time rather than fading. The authors' own reading was that people who benefited had kept doing the exercise on their own after the week ended, which makes it less a single intervention than the start of a habit.

One week of three good things, with a reason for each, was still measurable six months later. The people it helped most were the ones who quietly kept going.

Counting blessings

Two years earlier, Robert Emmons and Michael McCullough had published the experiment that started the modern gratitude literature. Across three studies, people kept weekly or daily records and were assigned to list either things they were grateful for, hassles that had annoyed them, or neutral events. The gratitude groups reported more positive mood, slept a little better, exercised a little more and, in one study, offered more help to others. The authors were careful: the gains showed on several measures, not all, and the clearest finding was on positive affect rather than on reducing anything negative. That distinction has held up ever since.

What the meta-analyses found

Single studies are where enthusiasm lives. Meta-analyses are where it gets audited. In 2016 Don Davis and colleagues pooled the gratitude intervention trials then available, in the Journal of Counseling Psychology, and found that against doing nothing, gratitude exercises improved psychological wellbeing by a small-to-moderate amount. Against an active comparison task, the advantage shrank to small for wellbeing and was not significant for anxiety. Their title, thankful for the little things, was a fair summary.

In 2021 David Cregg and Jennifer Cheavens asked a narrower question in the Journal of Happiness Studies: do gratitude exercises help with symptoms of depression and anxiety? Across 27 studies and 3,675 participants the answer was yes, slightly. The effect was small, it was smaller still against active controls, and the authors concluded that gratitude exercises should not be recommended as a standalone approach for someone struggling with those symptoms, which is a reasonable thing for anyone reading a wellbeing blog to hear plainly.

Real, and modest. Both halves of that sentence are the finding.

Specific, and not every day

Two details from the research change how the exercise works in practice. The first is specificity. Seligman's version asked for the reason behind each good thing, and that is where the effort is. Three items can be rattled off: coffee, the weather, a friend. Three items with causes require you to replay the day and notice the conditions that produced something good, often including something you did. That replay is plausibly where the effect comes from, and it is the part people drop when the exercise gets stale.

The second is frequency. In a 2005 paper on what makes happiness changes last, Sonja Lyubomirsky, Kennon Sheldon and David Schkade reported that people who counted their blessings once a week improved, while those asked to do it three times a week did not. Their explanation was that the daily version becomes a chore, and a chore done on autopilot stops producing the noticing that makes it work. Seligman's nightly week did not seem to suffer from this, so the honest conclusion is not a fixed schedule but a warning: the moment the exercise feels like a box to tick, do it less often and more carefully, not more often and faster.

On terrible days

Nothing in the literature says you must find three good things in a day that did not have them. Forcing gratitude over grief, illness or a genuinely bad week tends to produce a quiet resentment of the exercise, and possibly of yourself for failing at it. Emmons and McCullough's participants had a choice of small, concrete things, and the effect came from attention to what was there, not from pretending something was fine. On a terrible day, one good thing is allowed. So is none, with a plain note that it was a bad day, which is its own useful record. A gratitude habit that survives is one that was never asked to lie.

Doing it in Luvia

Luvia's journal has a gratitude field built into each entry, so the exercise sits next to the rest of the day rather than in a separate app you have to remember to open. You can attach photos, which turns out to be a good shortcut to specificity, because a picture of the thing makes it hard to write something vague about it. Entries autosave as you type, and they stay on your device; nothing you write is uploaded or read, which matters for a journal more than for most things, and is spelled out in the privacy notes. If you want the weekly rhythm the research hints at, do the gratitude field properly on Sundays and leave it blank the rest of the week without guilt.

That is the whole practice. Three things, a reason each, when it feels true. It will not rewire you. It might, quietly, make the ordinary days a little easier to see.