How many steps a day is good?

The short answer

About 7,000 steps a day. That is where the largest review of the evidence, published in The Lancet Public Health in 2025, found most of the benefit had arrived, across death from any cause, heart disease, type 2 diabetes, dementia, depression and falls. Ten thousand is a good day and still slightly better than 7,000. Four thousand is where the benefit starts to show. There is no official target: neither the World Health Organization nor the CDC recommends a step count, only minutes of activity a week.

The rest of this page is the evidence behind those three numbers, and the one thing the researchers keep saying about them: the specific figure should not get too much reverence.

Where 10,000 came from

Not from a study. In 1965 a Japanese company sold a pedometer called the manpo-kei, which means “10,000 steps meter”. The number was chosen because it sounded good, and it stuck for sixty years before anyone tested it properly. When they did, it turned out to be roughly right for younger adults and generous for older ones, which is a better record than most marketing numbers can claim.

What the research found

The 2025 review pooled 57 studies and more than 160,000 adults from over ten countries, followed for years with their steps measured by device. Against a baseline of 2,000 steps a day, 7,000 was associated with:

• 47% lower risk of death from any cause
• 25% lower risk of cardiovascular disease, and 47% lower risk of dying from it
• 37% lower risk of dying from cancer
• 14% lower risk of type 2 diabetes
• 38% lower risk of dementia
• 22% lower risk of depressive symptoms
• 28% lower risk of falls

For most of those outcomes the curve flattens between 5,000 and 7,000 steps: the risk keeps falling above that, but slowly. Heart disease is the exception, where the benefit kept growing past 7,000. The authors suggested 7,000 “may be a more realistic target than the current unofficial target of 10,000 steps per day”.

Three caveats the authors and the experts who reviewed the work make, and which belong next to the numbers. The cancer and dementia findings rest on a small number of studies and have low certainty. The people in these studies were generally healthy, so the figures do not transfer cleanly to anyone managing a chronic condition. And these are associations from watching people over years, not trials: walking more goes with living longer, and the studies cannot prove it is the walking that does it.

It changes with age

A 2022 meta-analysis of 15 cohorts and 47,471 adults found the point where more steps stop adding much: 6,000 to 8,000 a day for adults over 60, and 8,000 to 10,000 for younger adults. So 10,000 is a reasonable ceiling to aim at in your thirties and forties, and a good deal more than a seventy-year-old needs for the same benefit.

For the brain specifically, a 2025 study in Nature Medicine followed 296 older adults for up to 14 years with imaging for the proteins behind Alzheimer’s disease. Among those at risk, 3,000 to 5,000 steps a day went with cognitive decline delayed by about three years, and 5,000 to 7,500 by about seven, with no further gain above that. The mechanism was slower build-up of tau, and it needed only a modest count.

Is there a top?

Not one the studies could find. A 2023 meta-analysis of 226,889 people found risk of death kept falling with every extra thousand steps, about 15% per thousand, with benefits beginning at roughly 4,000 steps for all-cause mortality and 2,300 for cardiovascular mortality, and no level at which more became worse. The gains shrink as the count grows. They do not reverse.

How many steps people actually take

Fewer than any of these targets. The largest measurement, 68 million days of smartphone data from 717,527 people in 111 countries published in Nature in 2017, put the world average at 4,961 steps a day. The United States averaged 4,774, the United Kingdom 5,444, Japan 6,010, and Hong Kong, the highest, 6,880. Most adults are therefore in the tier below “low active”, and 7,000 is a real increase for most people rather than a rounding of what they already do.

Does it matter how fast?

Somewhat, and the step count does not know. Steps at a brisk pace of about 100 a minute count as moderate activity, the kind the WHO’s 150 minutes a week refer to, and several of the studies found faster steps went with a bigger benefit at the same total. Step counts also miss cycling, swimming and rowing entirely. If your exercise is not on foot, the numbers on this page understate what you do.

The same walk, counted toward somewhere

Seven thousand steps is about 5 km, an hour of walking spread across a day. The hard part is never the distance. It is doing it again tomorrow, and the week after, when nothing about the loop has changed. What keeps people going, in the words of the ones who do, is something to look forward to on the walk, progress they can see, and someone who notices.

That is what reTrek is built for. Pick a route, Everest Base Camp is free, and your daily count stops being a number and starts being distance along a real trail, with a short film about the place waiting at each milestone. Walk it with a friend and pool your steps, or race them to Base Camp and find out which of you skips fewer evenings.

The step tiers, and a page for each count

Steps a day	Tier	Pages
0 to 4,999	Sedentary	1,000, 1,500, 2,000, 2,500, 3,000, 3,500, 4,000, 4,500
5,000 to 7,499	Low active	5,000, 6,000, 6,500, 7,000
7,500 to 9,999	Somewhat active	7,500, 8,000, 8,500, 9,000
10,000 to 12,499	Active	10,000, 11,000, 12,000
12,500 and up	Highly active	13,000, 14,000, 15,000, 16,000, 17,000, 18,000, 19,000, 20,000, 21,000, 22,000, 23,000, 24,000, 25,000, 26,000, 27,000, 28,000, 30,000, 35,000, 40,000, 50,000

The tiers come from Tudor-Locke and Bassett's 2004 index for healthy adults, still the scale most step studies report against.