Ask almost anyone how many steps a day they should walk and the answer comes back instantly: 10,000. The number is on fitness trackers, magazine covers and NHS posters, and most people assume it came from decades of exercise science. It did not. It came from a Japanese advertising campaign in the 1960s, and the research that followed suggests the real health threshold is far lower.
Where the number actually came from
The 10,000-step target traces back to 1965, when the Japanese company Yamasa Tokei launched a pedometer it named manpo-kei. The name translates literally to “10,000 steps meter”. There was no clinical basis for the figure. It was chosen because it was a memorable, round number, the written character for 10,000 looked pleasing in advertising, and 10,000 steps sounded impressively ambitious to a postwar population that walked far more than that already without thinking about it. The device sold well, the number stuck, and decades later fitness tracker companies adopted it as a default goal because it was already in the culture.
In other words, one of the most widely believed health targets in the world began life as a marketing slogan. That alone does not make it wrong. But it does mean the number deserves the same scrutiny as any other popular health claim, and when researchers finally tested it, the results were surprising in both directions.
What the mortality studies actually found
The strongest evidence comes from large cohort studies that used accelerometers rather than self-reporting, because people overestimate their own activity badly. The headline findings are consistent across several independent studies:
- Most of the benefit arrives early. A 2019 study of older women in the US (I-Min Lee and colleagues, JAMA Internal Medicine) found mortality rates fell steeply from around 2,700 steps a day up to about 7,500, then flattened. Beyond 7,500, the study found no additional survival benefit.
- A 2021 Lancet Public Health meta-analysis pooling data from 15 international studies, around 50,000 people, found the mortality risk plateaued at roughly 6,000 to 8,000 steps a day for adults over 60, and 8,000 to 10,000 for those under 60.
- Every step still counts. The Lancet analysis found no threshold below which steps were worthless: each additional 1,000 steps a day was associated with a lower mortality risk across the whole range, including the first 2,000.
- Intensity matters less than believed. Walking pace made little difference once total daily steps were accounted for. A slow walker clocking 8,000 steps did about as well as a brisk one.
A widely covered 2023 analysis in the European Journal of Preventive Cardiology went further and found benefits starting at around 2,300 to 2,600 steps a day, with roughly an 11 percent lower cardiovascular mortality risk at 2,600 steps versus 2,000. That figure circulated with heavy caveats, though: it pooled observational studies of varying quality, and observational step data cannot fully separate the effect of exercise from the effect of being healthy enough to walk.
What the strongest trial evidence says
Observational studies show association, not cause. Fitter people walk more, and sick people walk less, so part of every reported benefit is reverse causation. The 2023 STEP trial published in the Lancet, which randomised people over 60 to a goal of 6,000-plus steps versus usual activity over roughly two years, remains the largest randomised test. It found no reduction in its primary combined endpoint of mortality and cardiovascular events, though participants did increase activity and report better quality of life. Defenders of the trial note that the average participant already walked over 7,000 steps a day at baseline, so the study asked whether adding steps to an already active life helps, not whether walking beats sitting. The trial was underpowered for deaths, and the follow-up was short.
The honest reading: the trials weaken the confident observational claims, but they do not overturn the basic case for walking more. The effect size claimed by the observational studies (roughly 50 percent lower mortality at 8,000 versus 4,000 steps) is almost certainly inflated. Something real is probably there, and it is plausibly closer to 10 to 20 percent than 50.
Does a step target even matter psychologically
There is one line of research suggesting the round number does more good than harm, and it deserves a fair hearing. A 2021 study in The Lancet Public Health found adults given a 10,000-step goal walked more than those given a placebo goal of 9,000 or an “anything is good” instruction. People appear to respond to concrete, ambitious round numbers, and the 10,000 target, despite its marketing origin, may motivate activity that a scientifically calibrated 7,500 target would not. The counterargument is that unreachable targets demotivate the least active people most, and they are the ones with the most to gain from 2,000 extra steps.
What each side actually claims
Those defending the 10,000-step standard say the number is roughly right for adults under 60, that observational data covering hundreds of thousands of people consistently shows lower mortality up to around 10,000 steps, and that a memorable target has population-level motivational value that a nuanced one never would. Those knocking it say the figure was invented to sell pedometers, that the survival curve flattens well below 10,000, that trial evidence for hard clinical benefit is weaker than the headlines suggest, and that presenting 10,000 as a minimum makes ordinary active lifestyles feel like failure.
The bottom line
Walking more is one of the best-evidenced health interventions available at zero cost. The 10,000 number specifically is a marketing artefact: the science points to a plateau somewhere between 6,000 and 8,000 steps for meaningful benefit, with gains starting from any increase at all and intensity mattering little. If 10,000 gets you out of the door, the number is harmless and probably helpful. If it feels unattainable, the evidence is firmly on your side that 4,000 to 6,000 well-short-of-the-poster steps captures most of what the poster number promises.