What the data actually shows
The clearest causal evidence comes from prosocial-spending experiments. Dunn, Aknin and Norton (2008, Science) found that people randomly assigned to spend a small sum of money on someone else reported greater happiness afterward than those assigned to spend the same amount on themselves. Because the helping was randomly assigned, this points to a genuine causal effect rather than just a correlation — though the amounts and effects are small.
Beyond money, decades of survey research link volunteering to greater wellbeing, life satisfaction, sense of purpose, and even lower mortality risk. The association is reasonably consistent across studies and countries. But much of this work is observational: people who volunteer differ from people who don't in ways (health, sociability, income, baseline mood) that are hard to fully separate out, so part of the relationship is selection rather than pure cause.
The pattern also fits a long-standing developmental idea. Erik Erikson described 'generativity' — caring for and contributing to others and to the next generation — as a central task of midlife, with stagnation as its opposite. Later research on generativity has tied it to greater meaning and wellbeing in middle and later adulthood, which lines up with what the prosocial and volunteering literatures find.
Why this feels different from how it actually is
Helping can feel like a cost rather than a source of meaning because the benefits are quiet and the demands are loud. The time, money, and effort are immediate and visible; the sense of significance accumulates slowly and is easy to overlook. So in the moment, helping often registers as one more thing to do, not as the thing quietly holding your sense of purpose together.
The culture frames meaning as something you find by looking inward or by achieving, which makes contribution feel secondary. We are encouraged to chase the calling, the goal, the personal breakthrough, while the more reliable, lower-drama source — being useful to other people — gets treated as a nice-to-have rather than a main ingredient.
There is also a real difference between helping that feels chosen and helping that feels coerced or depleting. The research is most positive about voluntary, autonomous helping. Obligation, burnout, and caregiving without support can carry the costs of helping without the meaning, which is part of why people's lived experience of 'helping' varies so much from the upbeat headline.
What the research says to do about it
Start small and concrete. The most solid evidence is for ordinary, low-cost acts — spending a little on someone else, doing a specific helpful thing — rather than grand commitments. The prosocial-spending work suggests it is the act of directing resources toward another person, not the scale, that does the work, which makes this unusually accessible.
Where you can, make helping social and relational rather than abstract. Helping specific people you can see benefit, or helping alongside others, tends to track better with wellbeing than impersonal giving. Volunteering through a group also layers in social connection, which is itself one of the strongest correlates of meaning.
Aim for sustainable, chosen contribution over heroic, depleting effort. The benefits show up most clearly when helping feels voluntary and within your capacity. Protecting against burnout — saying no, getting support when caregiving, keeping commitments realistic — is not selfish; it is what keeps helping a source of meaning rather than a source of strain.
What the research says does not help
Treating helping purely as a means to your own happiness tends to undercut the effect. The benefit appears strongest when the focus is genuinely on the other person; approaching a good deed mainly as a wellbeing hack is a weaker version, and people are fairly good at noticing the difference in themselves.
Waiting to help until you have found the perfect cause or have more time and money does not help. The evidence is that small, ordinary, present-tense acts carry much of the effect, so deferring contribution until conditions are ideal mostly defers the meaning along with it.
Giving past your limits does not multiply the benefit and often reverses it. Helping driven by guilt, obligation, or unsupported caregiving can produce exhaustion and resentment rather than meaning. More is not reliably better; chosen and sustainable beats large and depleting in the data.
Real numbers in context
The signature experiment is modest in scale, which is part of why it is persuasive. In Dunn, Aknin and Norton (2008, Science), people were given a small sum — around $5 to $20 — and randomly told to spend it on themselves or on someone else; those who spent on others reported higher happiness by the end of the day. The amounts were tiny, which suggests the effect is about the act of giving, not the price.
The volunteering evidence is broader but softer. Across many studies, volunteers report higher life satisfaction and sense of purpose, and some longitudinal work links volunteering to lower mortality risk — but because most of this is observational, the honest reading is 'reliably associated with' rather than 'proven to cause.' The direction is consistent; treat the precise magnitude with caution.