What counts as a wearable
A wearable is a small electronic device worn on or close to the body that measures something, shows something, or both. Most contain a few tiny sensors, a processor, a battery and a wireless radio, usually Bluetooth, that passes data to a phone. The category is wide: a basic step counter clipped to a belt and a headset that overlays graphics on a room are both wearables, but they have almost nothing else in common.
It helps to sort wearables by two questions. Where on the body does the device sit? And is it mainly measuring the wearer, or mainly presenting information to them? Placement determines what the sensors can see. Purpose determines how the device is designed, how it is powered and how much its readings should be trusted.
The sensors inside
A small set of components appears again and again.
- Accelerometers measure changes in speed along one or more directions. They are the basis of step counting, activity classification and fall detection.
- Gyroscopes measure rotation, which helps distinguish a swimming stroke from a wave of the hand.
- Optical sensors shine light into the skin and measure how much returns. Changes in that light follow the pulse and, in some designs, estimate blood oxygen.
- Electrical sensors pick up tiny voltage changes on the skin, used for heart rhythm recordings and some skin-conductance readings.
- Temperature sensors track skin temperature, which is not the same as core body temperature.
- Location receivers use signals from satellite navigation systems to record routes and pace outdoors.
None of these sensors measures “fitness” or “stress” or “recovery” directly. Those labels come from software that combines raw signals with assumptions about how bodies behave. Many of those assumptions are built using machine learning, which finds patterns in example data, so a metric is only as good as the examples and the fit between those examples and the person wearing the device.
Main categories
Step counters and fitness bands. The earliest mass-market wearables were pedometers, which counted steps with a mechanical or simple electronic motion sensor. Modern bands add pulse sensing and sleep estimates while staying slim and lasting a long time between charges. Their appeal is simplicity.
Smartwatches. A smartwatch is closer to a small computer on the wrist. It adds a bright display, notifications, apps, often a satellite receiver and a larger set of health features. The cost is battery life, which is commonly measured in days rather than weeks. How the pulse and motion readings actually work, and where they go wrong, is a topic of its own.
Smart rings. A ring puts sensors against the underside of the finger, where blood vessels are close to the surface and the device is less obtrusive overnight. A ring has very little room for a battery and no real screen, so it leans on a phone app. Many people choose one mainly for sleep and activity summaries. The same limits apply as for any consumer sleep estimate.
Chest straps and patches. A strap or adhesive patch near the heart can record the heart’s electrical signal more directly than a wrist sensor, which is why athletes often prefer straps for hard training. Some patches are prescribed by clinicians for short monitoring periods. That is a medical use and differs from a consumer gadget.
Continuous glucose monitors and other body sensors. Some devices use a tiny sensor under the skin to track glucose in the fluid between cells. These are health devices with specific medical uses, and decisions about them belong with a clinician.
Hearables. Earbuds can include motion sensors, microphones and sometimes pulse sensing. The ear is a stable spot, which helps with head movement tracking.
Head-worn displays. Smart glasses and headsets put information in front of the eyes. They raise different questions from fitness trackers, such as comfort, battery weight and bystander privacy. Fully immersive headsets and mixed reality devices are further along that spectrum.
Assistive and robotic wearables. Exoskeletons, prosthetics and brain-computer interfaces move beyond sensing toward replacing or augmenting a function. They are typically specialized and often clinical.
Why the numbers are estimates
Every wearable faces the same constraints: it is small, it moves around on a body, it runs on a tiny battery and it must be comfortable for hours. Skin tone, tattoos, hair, a loose strap, cold fingers and vigorous motion can all affect optical readings. A step counter can mistake pushing a stroller or typing for walking. For that reason, trends over weeks usually tell you more than a single reading, and a wearable is best thought of as a rough guide to patterns rather than a measuring instrument for diagnosis.
If a device appears to show something worrying, or if you have symptoms, the sensible next step is to speak to a qualified health professional rather than trust or dismiss the gadget.
Where the processing happens
Raw sensor data is plentiful, and sending all of it to a server would drain a battery quickly. So a wearable usually does some work locally, summarizes the result and syncs the summary to a phone or cloud account. This is a small-scale example of edge computing, where processing is done near where data is created. Devices that connect to other gadgets and services also form part of the wider internet of things.
Batteries and design trade-offs
Almost all wearables use small lithium-ion batteries, and battery size is the main constraint on design. A brighter screen, a constant location signal or continuous pulse sensing each use more power. Designers therefore make trade-offs. A ring gives up the screen to last longer. A watch accepts daily or every-few-day charging in exchange for features. Sampling less often saves power but can miss short events.
Data and privacy
Wearables generate a continuous record of where you are, how you move, when you sleep and how your pulse behaves. That is sensitive, and it is worth understanding who can see it, how long it is kept and whether it can be shared with other companies. These are questions of data privacy that go beyond any single device, and they are discussed in the context of body data in particular elsewhere on this site.
Choosing sensibly
A few general points apply to almost any purchase. Decide what you actually want to know before buying. Check which data leaves the device and what the account settings allow. Expect to charge it regularly. And remember that a more expensive device with more sensors is not automatically more accurate for the one thing you care about. A simple device used consistently often beats a complex one left in a drawer.