Article

iPhone Step Counter Accuracy: What Studies Show and How to Improve It

How accurate is the iPhone step counter? See what studies measured, why steps get undercounted or overcounted, and how to get more reliable counts every day.

← All articles

iPhone Step Counter Accuracy: What Studies Show and How to Improve It

Man walking on a treadmill with an iPhone

The iPhone step counter is accurate when you walk steadily with the phone on your body. It is much less reliable across a whole day, mostly because the phone is not always with you. In treadmill tests the average error stayed under 5%. In everyday life, studies found undercounts of roughly 12% to more than 20%. This guide covers what the research measured, what throws the count off, and what you can do about it.

Stepswidget
Keep Your Daily Steps in View
Steps Widget shows step progress on your Apple devices and sends personalized reminders to help you stay active without constant notifications.
Explore Steps Widget

Key Takeaways

  • In controlled treadmill walking, iPhone step counts landed within about 5% of manually counted steps on average.
  • In real life, the main error is missing steps: the phone counts nothing while it sits on a desk or in a bag.
  • Distance is a weaker number than steps. Studies found the iPhone’s distance estimate can run well above the real figure.
  • Carry the phone in a snug pocket, keep Apple Watch and iPhone sources in the right order in Health, and judge trends over weeks, not single days.

Table of Contents

How accurate is the iPhone step counter?

Under controlled conditions, very accurate. A Canadian validation study reported a mean bias under ±5% at treadmill speeds of 5, 7.5 and 10 km/h, compared with steps counted by hand. An earlier JAMA research letter found that smartphone apps differed from direct observation by between −6.7% and 6.2% on treadmills.

So for a person walking at a normal pace with the phone on their body, the sensor does its job. The trouble starts when you leave the lab.

What happens in everyday life

The same Canadian study compared iPhone counts with accelerometer-derived steps in free-living conditions. The mean bias rose to 21.5%, or 1,340 steps per day. The authors cautioned against using iPhones as research-grade pedometers, likely because the phones were not carried all the time.

A 2019 study from Japan, Amagasa and colleagues, found iPhone counts averaged 12% lower than a validated pedometer worn during normal days. A 2020 study of a smartphone and activity trackers in free-living settings reported that all devices generally undercounted total steps by more than 20% against a StepWatch reference.

The pattern is consistent. The sensor reads walking well, and the day-long total comes up short when the phone is not on you.

Setting What was measured Result
Treadmill, 5 to 10 km/h iPhone vs. manual count Mean bias under ±5%
Treadmill, smartphone apps Apps vs. direct observation −6.7% to 6.2%
Free living, Canada iPhone vs. accelerometer 21.5% mean bias (1,340 steps/day)
Free living, Japan iPhone vs. pedometer About 12% lower
6-minute walk, PAD patients iPhone steps vs. reference −7.2% ± 13.8%

The last row comes from a 2018 clinical study of people with peripheral artery disease, who often walk slowly and unevenly. Even so, the undercount was modest.

A caution about the evidence itself

Researchers do not agree on how to test these devices. The INTERLIVE network’s 2021 expert statement says many consumer-device validity studies were not evaluated rigorously or transparently, and it recommends common protocols. Comparison against a reference measure is a question every consumer sensor faces, from step counters to continuous glucose monitoring for weight loss. Treat any single percentage as a ballpark, not a promise.

Why the iPhone undercounts steps

The iPhone has no way to count steps it never felt. That sounds obvious, but it explains most of the gap you see in studies.

The phone was not with you

In the Japanese study, underestimation was largest among people who seldom carried their iPhones. It was also greater for people who usually kept the phone in a bag than for those who used a pocket. The trend across carrying-frequency groups was suggestive rather than conclusive (P = .08), but the direction matches common sense.

If you leave your phone on the kitchen counter while you cook, tidy up or walk to a colleague’s desk, those steps depend on another device, such as an Apple Watch, to be recorded.

A bag swings differently than a hip

A backpack or handbag moves out of step with your stride. A phone riding on your hip or in a front pocket moves with your body, which is what the motion sensor needs to recognize a step. If you carry your phone in a bag most of the day, expect a lower total than your real one.

Slow, shuffling or interrupted walking

Short bursts of movement around the house are the classic gap. Hartung and colleagues observed that accuracy depends heavily on activity type: steady walking is measured far better than intermittent movement and daily tasks with repetitive arm motion. If your day is a lot of pacing, standing and short trips, the total can drift further from the truth than on a long, steady walk.

Why the iPhone sometimes overcounts

Overcounting is rarer, but it happens, usually from noise rather than walking.

Oxford researchers working on step-counting algorithms found that loose carrying positions, such as a loose pocket, let the phone rebound and add signal noise that may cause overcounting. A tight pocket avoids much of that.

Arm movement matters on the wrist. The same Hartung study reported that wrist-worn monitors tended to misclassify arm movements as steps, while hip-worn smartphone data did equally well or better. If you see your Apple Watch run ahead of your iPhone, the watch may be picking up arm motion during chores, while the phone was simply not on you. Neither device is lying: they are measuring different things at different moments.

Steps vs. distance: why the miles look off

Some people find the step count believable but the distance suspicious. That is a known difference. In the peripheral artery disease study, the iPhone’s distance algorithm overestimated distance by 43% on average, largely because it overestimated stride length. The authors noted that custom algorithms may be needed for accurate clinical distance measurement.

The JAMA authors made a related point: many smartphone apps count steps well, but errors in hardware and software compound when steps are converted into distance or calories.

In practice:

  • Trust steps more than distance.
  • Treat calories as a rough guide only.
  • If distance matters to you, calibrate by walking a known route, such as a track lap, and compare.

What to do when your iPhone and Apple Watch disagree

When both devices record steps, Apple Health has to choose. According to Apple Support, Health prioritizes manually entered data first, then data from Apple devices, then data from apps and Bluetooth devices. You can reorder sources under Data Sources & Access.

What that means for you: if you wear a watch only part of the day, the combined total can include steps from both. If you want to dig into how the two devices combine and sync, our guide on how iPhone and Apple Watch count and sync steps walks through it, and Apple Watch steps accuracy and limitations covers the wrist side.

No watch? You can still get a solid count. See tracking steps on iPhone without an Apple Watch.

What to do when your iPhone and Apple Watch disagree

How to improve iPhone step accuracy

The most useful lever is simple: carry the phone more often. The Japanese study found less underestimation among people who carried theirs more frequently. The rest of this list follows from the evidence above.

  1. Keep the phone on your body. A front pocket or hip is better than a bag, based on the carrying-position findings.
  2. Use a snug pocket. A loose pocket can add noise and overcounting.
  3. Avoid judging single days. Compare your weekly average. Consistent undercounting shifts all days similarly, so trends stay meaningful.
  4. Check your sources. In Health, open Data Sources & Access and make sure the order matches how you actually wear your devices.
  5. Do a quick self-test. Walk 100 steps counting by hand, then compare. Repeat a few times with the phone in your pocket, then in your hand.
  6. Wear the watch if you have one. It covers the moments when the phone is across the room.

If you are wondering about walking on a treadmill desk, see whether walking pad steps count on iPhone and Apple Watch.

Does accuracy matter for your goals?

For most people, consistency beats precision. If your phone reads 10% low every day, a 8,000-step day is still clearly more active than a 4,000-step day. Use the number to compare yourself with yourself.

The real risk is a number so low that it hides your effort, or so unreliable you stop trusting it. Apple Health is a general wellness tool, not a medical device, so if you have a condition where exact step or distance measurement matters, ask your clinician what to use instead.

Our perspective

We built Steps Widget around a simple observation: the most common accuracy problem is not the sensor, it is the gaps. A phone left on a desk misses steps, and a count you rarely look at does little to change your day.

That is why Steps Widget shows your Apple Health steps right on the Home Screen, Lock Screen, Apple Watch and Mac. The count is in front of you wherever you are, so you notice when it looks too low and can move or pick up the phone.

It is also why our reminders are personalized. An on-device model learns your walking patterns and nudges you only when you are likely to miss your daily goal. You get a prompt when it helps, not a constant stream.

And because it all runs on your device, with no account and no data leaving your phone, you can track your movement without handing over a record of your day.

Zero

Make your steps visible with Steps Widget

Steps Widget reads your steps from Apple Health and displays them as widgets on your Home Screen, Lock Screen, Apple Watch and Mac. Its reminders are designed to arrive only when you are likely to miss your goal, and nudges are free for all users.

Stepswidget

The app is free for step tracking and reminders. The optional Customization subscription is $1.50 per month and unlocks extra widget styles and appearance options, including styles built from your own photos. Get Steps Widget and see your count without opening Health. If you want to compare layouts first, read about the best step counter widget for iPhone or how to see your steps without opening the Health app.

FAQ

Is the iPhone step counter accurate enough to trust?

For steady walking with the phone on you, yes. Treadmill studies found average errors under about 5%. Over a full day, studies found undercounts of around 12% or more, mainly because the phone is not always carried.

Why does my iPhone count fewer steps than my Apple Watch?

The watch is on your wrist all day, while the phone may sit on a desk or in a bag. The wrist can also register arm movement that the phone never feels. Check your source order in Health under Data Sources & Access.

Is the iPhone’s walking distance accurate?

Less so than steps. One clinical study found the iPhone’s distance algorithm overestimated distance by 43% on average, largely through overestimated stride length. Use steps as your main measure and calibrate distance on a known route.

How can I stop my iPhone undercounting steps?

Carry it more often, ideally in a snug pocket instead of a bag. Pair it with an Apple Watch if you have one, and judge your weekly averages instead of single days.

Sources