Does gratitude journaling actually work?

Ask whether gratitude journaling works and you get two kinds of answer. One says it rewires your brain, floods you with dopamine and cuts your stress. The other says it is toxic positivity in a notebook. Both are confident, and neither is describing the research.

Here is the short version, before the long one. Gratitude journaling has been studied, mostly in short trials with self-reported outcomes, and it produces modest positive effects that are clearest immediately after the practice and fade after it stops. Nobody has shown it makes you durably happier, treats anything clinical, or predicts what will happen to you personally. That is a smaller claim than the category makes, and it is still a good enough reason to spend two minutes a day on it.

The rest of this article is the working: three studies, what each one did, what it found, and what it cannot show.

The daily-list trial: Cunha et al., 2019

This is the study that comes closest to what people mean by gratitude journaling.

Brazilian adults were recruited online and randomly assigned to one of three conditions for 14 days. One group listed five things they were grateful for each day, looking back over the previous day. A second listed neutral events. A third listed hassles. Of 1,337 people randomized, 410 were included in the analysis.

What it found: immediately after the two weeks, the gratitude group reported higher positive affect than both other groups. At the 14-day follow-up, they were still higher than the hassles group, but no longer distinguishable from the neutral-events group. Outcomes other than positive affect were generally not distinguishable from the neutral-events control (Cunha et al., 2019).

What it cannot show. Four things, and they matter.

The attrition was severe: 927 of the people randomized stopped submitting their daily lists. The people who finished are, by definition, the people for whom this was tolerable, and they were a largely female, educated, self-selected online sample. Generalizing from them to you is a guess.

Everything measured was self-reported. Participants were writing down good things daily and then filling in scales about how positive they felt. That is not a neutral measurement setup, and no trial like this can be blinded — you always know which group you’re in.

The comparison group is the interesting problem. The neutral-events group frequently recorded events that were actually positive, which means the “neutral” control was partly a gratitude condition wearing a different label. That weakens the contrast, and it is the most likely explanation for the follow-up result washing out.

And the intervention was five items a day for 14 days, listed online for a study. It was not an app, not three items, not a long-term practice. What it supports is “in this study, over two weeks, this group reported higher positive affect.” Not “gratitude journaling improves mental health.”

What one recall exercise does: Layous et al., 2017

This one is narrower and, for our purposes, more interesting.

Across three experiments, participants wrote about a time someone did something they were grateful for. Their emotional response was compared against other writing exercises — some positive but not gratitude-related, some neutral.

What it found: the gratitude recall reliably produced more gratitude than the comparison exercises. It also produced more indebtedness and guilt (Layous et al., 2017).

That second finding almost never survives the trip from journal to blog post, and it should. It means the emotional payload of remembering something you’re grateful for is genuinely mixed. Thinking about someone who helped you can bring up a sense of owing them. Thinking about something good you were given can bring up a flicker of guilt about it.

If that happens to you, nothing has gone wrong. You have not failed at gratitude, you are not doing it with the wrong attitude, and it is not evidence that the practice is bad for you. It is a documented feature of gratitude recall. Any app or book that promises this practice only ever feels warm is either not looking, or not telling you.

What it cannot show: these were single-session lab exercises measuring how people felt immediately afterwards. They say nothing about a daily practice, nothing about weeks or months, and nothing about wellbeing. This is a study about the texture of one recall, not about outcomes.

Why lingering might matter: Langeslag & Surti, 2021

The third study is not about gratitude at all, and I include it because it is the honest source for a claim this category makes constantly.

Forty-nine undergraduates viewed positive and neutral images in a lab. For some, they were told to savor — to focus on the pleasurable aspects and try to sustain the positive feeling. For others, just to look.

What it found: savored images were rated more pleasant and more arousing at the time, and about twenty minutes later they were still rated more pleasant than the ones that had only been viewed (Langeslag & Surti, 2021).

What it cannot show: this was pictures, in a lab, with 49 students. It was not journaling and not gratitude. It does not show that writing an extra sentence about why something mattered causes savoring, that savoring improves long-term wellbeing, or that anyone who skips the extra sentence is missing out. It supports one carefully bounded idea: deliberately attending to the enjoyable part of something can intensify and briefly sustain the immediate experience of it.

That is why, in Grateful, the because is an invitation and never a requirement. The evidence does not license anything stronger than an invitation.

Is weekly better than daily?

You will run into a confident claim that weekly gratitude journaling beats doing it more often, because frequent practice makes the whole thing stale. It is repeated everywhere, usually with an explanation about adaptation attached.

It is worth knowing what it actually rests on. The best-known result compared people asked to journal once a week against people asked to journal three times a week, over six weeks, and found the once-a-week group reported greater gains in wellbeing. Note what that comparison was not: it was not weekly versus daily. Three times a week is an odd, artificial schedule, and it is a strange thing to generalize from.

Meanwhile daily-list studies do find effects, including the one above. The broader literature is not settled here, and it does not converge on a frequency the way the popular version implies. What studies in this area do broadly agree on is scale — most ask for somewhere between three and ten items per entry.

Treat “weekly beats daily” as contested, not established. For what it’s worth, our own reasoning for a daily practice is not an evidence claim at all: a daily action is far easier to make automatic than a weekly one, and a practice you don’t do is worth nothing regardless of what the optimal frequency turns out to be. The staleness problem that the weekly recommendation is trying to solve is real, but it is a problem about running out of things to notice, and slowing down how often you write is only one possible answer to it.

So what does the evidence actually support?

Stacking up what those three studies can carry:

• Gratitude practices have been studied as a way to notice positive parts of everyday life, and short daily-list practices have produced higher self-reported positive affect immediately afterwards.
• The effects observed are modest, measured mostly against imperfect control conditions, and they fade once the practice stops.
• Recalling something you’re grateful for changes how you feel in the moment, in a way that includes uncomfortable feelings as well as good ones.
• Deliberately dwelling on something enjoyable can extend the pleasant experience of it, at least for a while.

What the evidence does not support, in either direction: that gratitude journaling treats or prevents any clinical condition, that it will make you happier, that it changes your brain in any way anyone has measured in a journaling context, that specific entries beat general ones, or that a particular number of items or words is optimal. Every one of those is claimed routinely. None of them holds up.

And there is a structural point that the individual studies don’t capture. This is a small literature of short trials, with self-report outcomes, heavy attrition, and control conditions that are difficult to design because you cannot give someone a placebo journal. Those are the conditions under which small effects are easy to find and hard to trust. Being sceptical of the enthusiastic version is reasonable.

The honest answer

Does gratitude journaling work? It depends what you were hoping “work” meant.

If you were hoping for a treatment, no. The evidence is nowhere near that, and if you’re dealing with depression or anxiety, this is not the intervention and no notebook is.

If you meant: is there decent reason to think spending two minutes a day writing down specific good things has a modest, mostly short-term positive effect on how you feel? Then yes, roughly. That is what the research supports, and it is not nothing.

But the reason I do this, after ten years, is simpler and doesn’t need a citation. It costs two minutes. It is pleasant while you’re doing it. And it puts your attention somewhere useful for a moment each day, in a life where attention otherwise defaults to what’s unfinished and what’s coming. You don’t need a randomized trial to justify that any more than you need one to justify taking a walk.

That’s a small enough claim that you can just try it and see. How to start takes about five minutes to read, and what to write is the part people actually get stuck on.

Grateful is the app we built on exactly this reading of the evidence — modest about what it claims, and designed so the practice is easy to keep rather than easy to fail.