Is a pedometer accurate at all?
The short answer: accurate enough for tracking trends, but not always accurate enough to count every single step. No everyday pedometer directly sees the step itself. It detects acceleration through motion sensors, then an algorithm decides whether the movement pattern looks like walking.
The main takeaway: a pedometer is more useful as a consistent yardstick for comparing your own days than as an absolute counter you can trust down to the exact step.
In the 2024 study, these figures were obtained by comparing devices with an ActivPAL accelerometer, which was used as the reference. This does not mean that every Apple Watch will be off by 6,4%, or every smartphone by nearly 30%. The result depends on the model, where you wear the device, walking speed, stops, and whether you carry your phone with you all day.
- For daily motivation, consistent conditions and one data source matter more.
- When comparing two devices, the brand is not the only factor—the place where you wear it matters too.
- For medical or research measurements, an everyday gadget is not enough without separate validation.
Why one day can produce three different numbers
A smartphone, watch and fitness tracker may count the same walk differently because they are located on different parts of your body. A phone usually sits in a pocket and senses body movement. A watch and fitness tracker sit on your wrist and also respond to arm movements. Each device therefore receives its own set of signals and applies its own algorithm.
| Device | What it detects | Main strength | Where it most often fails |
|---|---|---|---|
| Smartphone | Body movement in a pocket or bag | No need to buy a separate device | The phone is left at home, sits in a backpack or does not sync for a long time |
| Smartwatch | Wrist and body movement | Usually always on your wrist and shows data quickly | Your arm is still, steps are slow, or you are pushing a cart or holding onto support |
| Fitness tracker | Wrist movement, sometimes body movement | Lightweight and convenient for continuous wear | Using your hands, washing dishes, a loose fit or an unusual gait |
That is why there is no universal answer to the question “Which device is the most accurate?” For a walk where you naturally swing your arms, a watch may work well. For someone pushing a stroller or holding a handrail, a wrist-worn device may miss steps. A smartphone that stays in your pocket all the time may sometimes be more practical than a watch you take off for several hours.
Where missed and extra steps come from
When steps are added
Extra steps appear when a sensor detects a characteristic rhythmic movement even though you are not walking. This happens especially often with wrist-worn devices: the algorithm sees arm movements and interprets them as a walking rhythm.
- washing and drying dishes;
- cleaning, vacuuming and active arm movements;
- riding on a bus or in a car over a bumpy road;
- moving on an exercise bike or rowing machine;
- gesturing and repetitive hand movements at your desk;
- using several apps at the same time that process the same signal differently.
When steps are missed
Missed steps are more likely when a step looks too weak or unusual to the algorithm. A device may fail to recognize short movements around a room, slow walking, the first steps after stopping, or movement without the usual arm swing.
- you walk very slowly or take small steps;
- you hold a handrail, cane, walker or cart;
- you push a stroller and barely move your arm;
- your phone is in a bag or backpack, or remains on a table;
- your route includes frequent stops and turns;
- you walk on stairs or an incline, or after changing your usual gait.
With slow walking, the error can increase noticeably. In a separate study, Fitbit Ultra undercounted steps at the lowest speed and did not record any steps at all during some one-minute intervals for some participants. If you are recovering from an illness, injury or surgery, do not rely only on the number in the app; also consider how you feel, how long you walk and how well you tolerate the load.
A 15-minute home accuracy test
You do not need a treadmill or a laboratory accelerometer. For an everyday check, all you need is a manual count, a level route and three repetitions. A hallway, school track or quiet path without stairs and traffic lights is best.
- Check the settings. Make sure the app has permission to access motion, and that the device is charged and synced.
- Record the starting values. If you cannot reset the counter, write down the number before starting the test.
- Put on your watch or fitness tracker as you usually wear it. Put your smartphone in your usual pocket, not in your hand.
- Start counting steps manually. Counting aloud or using a mechanical clicker is convenient.
- Walk exactly 500 steps at your usual pace. Do not actively gesture while talking, stop or change the position of the device.
- Record the result from each device. For the phone and watch, use data from the same time interval.
- Repeat the test two more times. One random missed step or extra movement should not determine the conclusion.
- Then repeat a short test at a slow pace or with frequent stops if that is how you usually walk.
Calculate the error as follows: (device reading − manual count) ÷ manual count × 100%. For example, if you count 500 steps manually and the device shows 470, the error is −6%. A minus sign means undercounting; a plus sign means overcounting.
For personal use, you can use this practical guide: up to 5% is a very good result, 5–10% is suitable for tracking trends, and more than 10% is a reason to check the fit, the phone’s position and your walking style. This is not a universal medical threshold. The INTERLIVE expert recommendations propose an error level of up to 5% as a strict benchmark for device validation, but for an everyday gadget, consistency in your usual conditions matters more.
Which number should you trust each day?
If after three repetitions one device consistently shows 480–490 steps for 500 manual steps, while another shows 430 one time and 540 the next, the first is better suited to your personal monitoring. Even if its absolute result is not perfect, it will be useful as a consistent scale.
- There is no universally most accurate type of device: the result depends on the model, algorithm and where you wear it.
- Watches and fitness trackers may add steps because of arm movements, while a smartphone may miss them if you do not keep it with you.
- Slow walking, frequent stops, stairs, a stroller and holding onto a handrail increase the chance of error.
- Test your gadget with three repetitions of 500 steps and separately assess your usual pace.
- For tracking progress, use one device, one wearing position and consistent conditions instead of comparing figures from different ecosystems.
How to track walking without unnecessary anxiety
- Choose one main step source. If you have both a watch and a phone, do not add their values together manually.
- Wear the device consistently: on the same wrist, in the same pocket or on the same strap.
- Look at the weekly average rather than a difference of 100–300 steps between two days.
- Note the context: an outdoor walk, workday, trip, workout or household chores.
- If you change devices, start a new observation period and do not compare it directly with the old one without testing.
- To assess the walking itself, complement steps with walking time, pace and how you feel.
If you are choosing a goal, you do not have to turn it into a competition with a perfect number. In the articles how many steps you really need and what the 10,000-step goal means, consistency and gradually increasing activity matter more. And if you want to assess intensity, add walking cadence and pace to your step count.
Which is more accurate: a smartphone or a smartwatch?
On average, there is no winner for all models. If your phone stays in your pocket, it may track walking well. If you wear your watch all day and naturally move your arms, it is often more convenient. In the 2024 study, the watches tested showed a smaller error than smartphone apps, but that result applies to specific models and conditions.
Why does my watch count steps when I wash dishes?
The wrist sensor detects repetitive arm movements and may interpret them as a walking rhythm. This is a known limitation of algorithms: similar false steps have been observed during household tasks, travel and other movements without walking.
Can I add the steps from my phone and watch together?
No. Both devices may count the same walk, so the total will almost certainly be too high. Choose one main source or use an app that manages data priority and removes duplicates automatically.
What should I do if I walk slowly or push a stroller?
Test the device in exactly that situation. With slow walking and no arm swing, some steps may be missed. In this case, it is also useful to track walking time, route and how you feel rather than relying only on the step count.
Do I need to replace my device if it is off by 10%?
Not necessarily. If the error is consistent and you compare your days using the same device, it is still useful. Consider replacing it if the result is unstable, the device often loses data or it does not suit the way you walk.
Sources
- Case MA, Burwick HA, Volpp KG, Patel MS. Accuracy of Smartphone Applications and Wearable Devices for Tracking Physical Activity Data. JAMA, 2015. DOI 10.1001/jama.2014.17841
- Hong K-R, Hwang I-W, Kim H-J et al. Apple Watch 6 vs. Galaxy Watch 4: A Validity Study of Step-Count Estimation in Daily Activities. Sensors, 2024. DOI 10.3390/s24144658
- Johnston W et al. Recommendations for determining the validity of consumer wearable and smartphone step count: expert statement and checklist of the INTERLIVE network. British Journal of Sports Medicine, 2021. DOI 10.1136/bjsports-2020-103147
- Fuller D et al. Reliability and Validity of Commercially Available Wearable Devices for Measuring Steps, Energy Expenditure, and Heart Rate: Systematic Review. JMIR mHealth and uHealth, 2020. DOI 10.2196/18694
- Germini F et al. Accuracy and Acceptability of Wrist-Wearable Activity-Tracking Devices: Systematic Review of the Literature. Journal of Medical Internet Research, 2022. DOI 10.2196/30791
- Wong CK, Mentis HM, Kuber R. The bit doesn't fit: Evaluation of a commercial activity-tracker at slower walking speeds. Gait & Posture, 2018. DOI 10.1016/j.gaitpost.2017.10.010
- Park W, Lee VJ, Ku B, Tanaka H. Effect of walking speed and placement position interactions in determining the accuracy of various newer pedometers. Journal of Exercise Science & Fitness, 2014. DOI 10.1016/j.jesf.2014.01.003
Count your steps with Qozgal
A free app that counts your steps, keeps your streak and motivates you to walk every day.