How many steps a day you actually need: not 10,000

You are standing outside your building, staring at your phone. 8,400 steps. You are about a kilometre short of the round number, and that familiar feeling creeps in: the day does not count. Sound familiar?

Here is the good news. Since July 2025 you can stop doing that math. The largest review of step-count research to date found that most of the benefit is banked at around 7,000 steps a day, and the curve flattens after that. The 10,000 figure that has made millions of people feel lazy was never a scientific standard. It was an advertising slogan for a Japanese pedometer.

Below: where the 10,000 came from, what the current numbers say, why the target is different at 65 than at 35, and what to do if you sit all day. That last one comes with an uncomfortable detail, and we will get back to it near the end.

Where 10,000 steps came from

A marketing department invented 10,000 steps, not a laboratory. In 1965 the Japanese clock company Yamasa launched the first mass-market pedometer and called it manpo-kei, which translates as ten-thousand-step meter.

The timing was perfect. Tokyo had just hosted the 1964 Olympics, the country was caught up in fitness, and a round, memorable goal fit the mood. The Japanese character for ten thousand looks good, reads fast, and sat neatly on the device dial. No research sat behind the number at all.

Then it snowballed. Fitness bands picked it up, then apps, then corporate wellness programmes. By the 2010s 10,000 had become a standard nobody questioned, because everyone assumed somebody else had already checked it. Researchers noticed and started counting.

What the 2025 evidence says

Roughly 7,000 steps a day is the point where extra benefit largely stops. That comes from a systematic review published in The Lancet Public Health in July 2025. A team led by Professor Melody Ding at the University of Sydney pooled 57 studies run between 2014 and 2025 across more than ten countries, including Australia, the United States, the United Kingdom and Japan. The quantitative analysis covered more than 160,000 adults.

A systematic review with meta-analysis means the researchers ran no new experiment. They collected every solid study on the question and calculated a combined result. That carries more weight than any single study, because the quirks of individual samples cancel each other out.

Comparing people who walk 7,000 steps with people who walk 2,000, the review found:

• risk of death from any cause lower by 47%
• cardiovascular disease lower by 25%
• dementia (loss of memory and thinking ability in later life) lower by 38%
• depression lower by 22%
• falls lower by 28%
• type 2 diabetes lower by 14%
• cancer lower by 6%

Look at the spread. Dementia and mortality respond strongly to walking, cancer barely moves. That matters more than it seems. Walking is not a universal pill, it acts on specific systems, and it is more honest to talk about the heart, the blood vessels and the brain than about health in general.

As for 10,000: on dementia, the gap between 7,000 and 10,000 steps was about 7 extra percentage points of risk reduction. For heart disease the benefit did keep climbing past seven thousand. For most other outcomes the graph reached a plateau, meaning each additional thousand steps bought less and less.

What happens at each thousand steps

The biggest jump in benefit happens at the very start, not at the finish. This is probably the most useful finding in the whole field, and the one most people miss.

Here is what the research shows at each mark:

• 2,000 steps is the baseline in most studies. That is roughly a person who stays home or at a desk and barely goes out.
• 3,000 to 4,000 steps already shows a clear benefit. The 2025 review states plainly that even modest counts around 4,000 steps are linked to better health than 2,000.
• 4,000 to 4,500 steps delivers about half of the total available risk reduction. That figure comes from an analysis of 72,174 British participants in the UK Biobank project, published in the British Journal of Sports Medicine.
• 7,000 steps is where most of the benefit has been collected and the curve visibly flattens.
• 9,000 to 10,000 steps gave a 39% lower risk of death and a 21% lower risk of heart disease in that same British analysis. Real, but nothing like the gain from moving between two thousand and four thousand.

Think of filling a glass from a tap. The first seconds deliver almost all the water, the last drops add a millimetre. Someone walking 2,000 steps who adds another 2,000 gains more than someone walking 8,000 who pushes to 10,000. The second version looks more athletic. The first buys more health.

That flips the usual logic. A person on 3,000 steps looks at the 10,000 target, decides it is out of reach, and does nothing. Yet one lap around the block would do more for them than an extra kilometre does for a committed walker.

Why the target drops after 60

Age changes the number, and it changes it a lot. A meta-analysis of 15 international cohorts (groups of people followed for years) published in The Lancet Public Health in 2022, led by Amanda Paluch of the University of Massachusetts Amherst, covered about 50,000 people across four continents.

The answer split by age:

• for adults 60 and older, the risk of early death stopped falling at roughly 6,000 to 8,000 steps a day
• for adults under 60, at roughly 8,000 to 10,000

Earlier, in 2019, Harvard epidemiologist I-Min Lee published a study in JAMA Internal Medicine covering 16,741 women with an average age of 72. Women walking 4,400 steps a day were 41% less likely to die during the follow-up period than those walking 2,700. The benefit kept rising to about 7,500 steps, then the line went flat.

The logic is simple. The older the body, the smaller the load needed to produce an effect. For a 72-year-old, a walk to the shop and back is genuine cardiovascular training. For a 30-year-old office worker the same walk changes almost nothing, because their heart is already used to it.

There is a practical takeaway for anyone pushing their parents into fitness apps. Older people do not need ten thousand. They need a steady six or seven thousand and no sense that they failed an assignment.

So much for quantity. Next comes a question the researchers themselves disagree on: does speed matter?

Does walking fast matter?

The data conflict here, and it is better to show that than hide it. Two strong studies reached different answers.

The Paluch meta-analysis found no link between pace and mortality once total steps were accounted for. The authors put it directly: what mattered was getting the steps in, not how fast they were walked.

An analysis of 78,500 British adults aged 40 to 79, published in 2022 by researchers from the University of Sydney and the University of Southern Denmark, found the opposite. There, a faster pace added benefit on top of the step count. The same work linked 9,800 steps a day to roughly half the risk of dementia, and 3,800 steps to about a quarter less risk.

So what should you do with a split like that? Treat pace as a bonus, not a requirement. If you can walk hard enough that your breathing picks up, you will probably get a little more out of it. If you cannot, the steps still count. Nothing is cancelled because you walk slowly with a pram or with a dog that stops at every bush.

What if you only walk at weekends?

Rare but solid walking days appear to work too. This may be the most encouraging finding for anyone whose schedule leaves no room for daily strolls.

Researchers from Kyoto University and the University of California, Los Angeles analysed 3,101 American adults with an average age of 50. Each wore an accelerometer (a motion sensor that counts steps more accurately than a phone in your pocket) for one week, and their health was tracked for the next 10 years. The study appeared in JAMA Network Open in 2023.

People were grouped by how many days a week they hit at least 8,000 steps:

• zero days: the baseline
• 1 to 2 days a week: mortality lower by 14.9%
• 3 to 7 days a week: mortality lower by 16.5%

The gap between a couple of days and almost every day was under two percentage points. A person who only walks on Saturday and Sunday gets nearly as much as the daily walker.

This is an observational study, and the authors list the weak spots honestly: activity was measured for a single week, and people wearing a tracker often move more than usual simply because they are being watched. Still, the direction matches other research on weekend exercisers.

Now for that detail about sitting.

Desk job: do steps cancel out the sitting?

Steps offset part of the harm from sitting, but not all of it. An analysis of 72,174 UK Biobank participants (average age 61, followed for about seven years) was published in the British Journal of Sports Medicine.

One number stands out: participants spent an average of 10.6 hours a day sitting or lying down. Not a workday, a full daily total including the sofa and the commute.

The finding: any increase in steps above roughly 2,200 a day lowered risk both for people who sat little and for people who sat a great deal. An office worker with ten hours at a desk still gains from walking rather than having already ruined things.

One of the authors, Dr Matthew Ahmadi, framed the result without illusions: this is by no means a get out of jail card for long hours of sitting, though all movement matters. Steps work like an airbag, not like cancelling the crash.

In practice: if you work at a desk, breaking walking into 10 to 15 minute pieces through the day beats sitting for nine hours and heroically covering five kilometres at night. Muscles and blood vessels want regular activation, not one long act of penance. If you are building up load and hitting fatigue, it helps to understand how the body actually recovers.

How much the world actually walks

The average person on the planet walks about 5,000 steps a day, well below any of the recommended figures. The data come from a Stanford University team: 68 million days of activity from 717,527 people in 111 countries, collected through smartphones and published in Nature in 2017.

The spread between countries is enormous, and it is explained by city design more than by willpower. The same study found that in cities where walking is easy, the gap between the most and least active residents is smaller. People walk where there is somewhere to walk to.

A current picture comes from a pooled analysis of 507 surveys covering 5.7 million participants in 163 countries, published in The Lancet Global Health in 2024. In 2022, 31.3% of adults failed to reach the minimum activity WHO recommends, up from 23.4% in 2000. By region the differences multiply:

• high-income Asia Pacific (Japan, South Korea, Singapore): 48.1% insufficiently active
• South Asia: 45.4%
• Central Asia, Middle East and North Africa: 38.5%
• Latin America and the Caribbean: 36.6%
• high-income Western countries: 27.7%
• East and Southeast Asia: 24.6%
• Central and Eastern Europe: 22.7%
• Sub-Saharan Africa: 16.8%

Notice the paradox. The richest, most technological regions move the least, and Sub-Saharan Africa moves the most. Where there are fewer cars, lifts and delivery apps, the body moves by default, no app required.

Kazakhstan has its own national study: 6,720 adults surveyed across all 17 regions using the WHO method, published in 2025. There, 19.7% of adults fall short of the recommendations, 21.6% of women and 17.8% of men. Inside the country the range is wide: in Aktobe 37.3% are insufficiently active, while in East Kazakhstan 67.6% count as highly active. Among its neighbours the country sits in the middle: Kyrgyzstan 11.4%, Uzbekistan 16.4%, Tajikistan 28.3%.

What these numbers do not prove

All of this research shows association, not cause. The distinction is fundamental, and good scientists insist on it themselves.

These are observational studies: nobody was randomly assigned to walk or not walk, researchers simply watched how people already live. That leaves two traps:

• Reverse causation. People may walk less because they are already ill and find it hard, not the other way around. Then the illness is the cause, not the low step count.
• Hidden differences. People who walk a lot tend to eat better, smoke less, and have more money and free time. Part of the effect may belong to those factors.

Independent experts who reviewed the 2025 analysis for the UK Science Media Centre add detail. Dr Daniel Bailey of Brunel University London notes that several outcomes rest on a small number of studies, and that participants were generally healthy people, so the findings do not transfer directly to someone living with a chronic condition. Professor Steven Harridge of King's College London points out that the review shows the link without explaining the mechanism behind it. Dr Andrew Scott of the University of Portsmouth reminds readers that a step count is a blunt measure: it misses intensity and barely registers swimming, cycling or rowing.

The review's own authors acknowledge that evidence for cancer and dementia rests on few studies with low certainty, and that some of the included work did not adjust for age or frailty.

None of this makes the result useless. It makes it a guide rather than a law of physics.

How to add 2,000 steps without changing your life

Two thousand steps is roughly 15 to 20 minutes of easy walking, and it fits into a normal day more easily than it sounds. No gym membership, no dawn heroics.

What actually works:

• get off one stop early on the way home (about 800 to 1,200 steps)
• walk around the flat or the corridor during phone calls (half an hour of calls is 1,500 to 2,000 steps)
• park at the far end of the car park instead of by the door
• take the stairs to the third or fourth floor, then the lift the rest of the way
• buy groceries in small trips instead of one big weekly shop

One more thing: do not mistake caffeine buzz for real energy. A 20 minute walk gives a steadier lift than an energy drink, which comes with its own bill for the body.

Back to that doorway and the 8,400 on the screen. By the 2025 evidence, that day was counted long ago. You collected the bulk of the benefit somewhere around the seventh thousand, before you even reached for your phone.

Your two-minute step for today: open the step counter on your phone, look at last week's average, and add 1,000 to it. That is your target for the next seven days. Not ten thousand. Your own number plus a thousand.

Sources

• University of Sydney: Ding et al., The Lancet Public Health (2025)
• The Lancet Public Health: press release on risk reductions at 7,000 steps (2025)
• American College of Cardiology: journal scan of the meta-analysis (2025)
• Science Media Centre: independent expert reaction (2025)
• Paluch et al., The Lancet Public Health: 15 international cohorts (2022)
• Lee et al., JAMA Internal Medicine: 16,741 women, 4,400 steps (2019)
• Ahmadi et al., British Journal of Sports Medicine: sitting and steps, UK Biobank
• Inoue et al., JAMA Network Open: 8,000 steps on 1 to 2 days a week (2023)
• Althoff et al., Nature: 717,527 people, 111 countries (2017)
• The Lancet Global Health: 507 surveys, 5.7 million participants (2024)
• National physical activity study in Kazakhstan (2025)
• WHO: physical activity recommendations and global statistics