Biometric Wearable 101: What These Sensors Actually Measure and How

Most people who buy a biometric wearable have no idea what the device is actually doing on their wrist. They see a number (a heart rate, a sleep score, a recovery percentage) and trust it. But understanding what the sensors are actually doing, and what they can and can't genuinely tell you, makes that data far more useful.

This is a plain-language breakdown of the core sensors found in biometric wearables in 2026: what each one measures, how it works, and what the numbers actually mean.

The Optical Heart Rate Sensor (PPG)

Almost every wearable uses photoplethysmography, or PPG. The sensor shines a light (usually green LED, sometimes infrared) into the skin on the underside of the device. Blood absorbs light differently depending on how much is flowing through the vessels at any given moment, and the sensor reads those fluctuations to produce a pulse rate.

That's your heart rate reading.

PPG is reliable at rest and during steady-state activity. It becomes less accurate during high-intensity movement because motion artifacts, the sensor shifting slightly against the skin, introduce noise. Most wearables use accelerometer data to filter some of that out, but it's not a perfect fix.

One thing PPG cannot do: it doesn't measure the electrical signals from the heart. For that, you need ECG, which is a separate sensor entirely.

Heart Rate Variability (HRV)

HRV is one of the most discussed and most misunderstood metrics in consumer health tracking.

Your heart doesn't beat at perfectly even intervals. The gap between beats varies slightly from one beat to the next. That variation is HRV. Higher HRV generally indicates that your autonomic nervous system is handling stress and recovery well. Lower HRV, especially relative to your personal baseline, often signals fatigue, illness, or accumulated stress.

HRV is measured using the same PPG sensor as heart rate, but the calculation is more demanding. The device needs to detect the precise timing of each beat, then calculate millisecond-level differences between consecutive ones. This is why HRV readings are most accurate at rest, typically during sleep or first thing in the morning.

Single-point HRV readings are noisy. The signal becomes genuinely useful when you look at trend data across days and weeks.

Accelerometer and Gyroscope

Every biometric wearable includes a motion sensor. The accelerometer detects linear movement in any direction; the gyroscope detects rotation. Together, they tell the device how you're moving, how fast, and with what intensity.

That data feeds several downstream calculations:

• Step count is derived directly from accelerometer patterns
• Calories burned combines motion data with personal metrics (weight, age, height) alongside heart rate to estimate energy expenditure
• Sleep staging uses movement patterns and heart rate to distinguish light sleep, deep sleep, and REM phases
• Activity type is inferred from motion signatures, a run looks different from a walk, which looks different from cycling

The accelerometer also improves continuous heart rate monitoring. When the device detects movement, it applies correction algorithms to reduce PPG noise.

Sleep Staging

Sleep tracking in consumer wearables runs primarily on two inputs: movement and heart rate, including HRV patterns. The device uses those signals to classify sleep into stages.

Here's what each stage reflects:

• Light sleep (N1/N2): Reduced movement, slower and more regular heart rate. The transition phase in and out of deeper sleep.
• Deep sleep (N3/slow-wave sleep): Very little movement, lowest heart rate, lowest HRV variability. The most physically restorative stage.
• REM sleep: Low movement, the body is largely paralyzed, but heart rate and HRV become more variable again, closer to waking patterns. This is where most dreaming occurs and where memory consolidation happens.

Consumer wearables can't match the accuracy of a clinical polysomnography study. But for tracking your own patterns over time: whether your deep sleep is shrinking, whether alcohol disrupts your REM, whether a new training block is affecting recovery. They're genuinely useful.

VO2 Max Estimation

VO2 max is the maximum rate at which your body can consume oxygen during intense exercise. It's one of the strongest predictors of long-term cardiovascular health and aerobic fitness.

Consumer wearables don't measure VO2 max directly. They estimate it, typically by modeling the relationship between your heart rate and your pace or exertion level during sustained aerobic activity. The device builds a picture of how hard your heart works relative to how much work you're doing, then extrapolates a VO2 max figure from that.

The key word is estimation. These numbers are directionally useful. Tracking whether your estimated VO2 max improves over a training cycle is meaningful. Treating the absolute number as clinically precise is not. A lab-based test using a metabolic analyzer is the only way to get a validated measurement.

Skin Temperature Sensing

Some wearables include a skin temperature sensor, typically a thermistor. It measures the temperature of the skin surface, not core body temperature. The two are different.

Skin temperature is most useful as a relative indicator. Significant deviations from your personal baseline, particularly overnight, can signal illness, hormonal changes, or elevated physiological stress. Several devices use overnight skin temperature trends as an input to their recovery and readiness scores.

ECG (Electrocardiogram)

ECG measures the electrical activity of the heart directly. Unlike PPG, which infers heart rate from blood flow, ECG detects the electrical signals that cause the heart to contract.

Consumer ECG in wearables typically requires an active reading. You touch a sensor on the device to complete an electrical circuit. It's not running passively all day. The primary use case is detecting irregular rhythms, particularly atrial fibrillation (AFib).

Devices like the Apple Watch Series 11 and Withings ScanWatch Nova include ECG. It's a meaningful feature for users who have a clinical reason to monitor for arrhythmias. For general fitness and recovery tracking, most people won't interact with it regularly.

Blood Oxygen Saturation (SpO2)

SpO2 sensors use red and infrared light to estimate the percentage of hemoglobin in your blood that's carrying oxygen. Healthy resting SpO2 typically falls between 95 and 100 percent.

Consumer SpO2 readings are useful for flagging potential sleep apnea, repeated overnight dips are a common indicator, and for tracking how altitude affects your physiology. They're not medical-grade pulse oximeters and shouldn't be used for clinical monitoring.

How the Data Becomes Useful

Individual sensor readings are noisy. A single heart rate number, a single HRV reading, a single sleep score, none of these tell you much on their own.

The value of a biometric wearable comes from synthesizing multiple data streams over time. When heart rate, HRV, movement, sleep staging, and other inputs are combined and viewed as a pattern, they start to reflect something real about your physiological state.

This is why how a device presents data matters as much as what it measures. A wearable that surfaces raw numbers without context puts the interpretive work on you. One that synthesizes those inputs into a coherent picture of your health on any given day is more immediately useful for most people.

Hiyd takes this approach with a companion app feature called The Reading: a single descriptive view that consolidates heart rate, HRV, sleep staging, movement, calories, VO2 max estimation, and nutrition data. Rather than issuing prescriptive alerts or coaching nudges, it presents what the data shows. That distinction matters if you want to understand your body rather than be managed by an algorithm.

What No Sensor Can Tell You

Even the best wearable sensor array has real limits.

No consumer wearable can diagnose a medical condition. A low HRV trend doesn't mean you're sick. A poor sleep score doesn't mean you have a sleep disorder. An irregular heart rate reading may or may not be clinically significant.

Wearable data is most useful as a personal baseline tool. Your numbers compared to your own history, across weeks and months, are far more meaningful than your numbers compared to population averages or manufacturer benchmarks.

The sensors are measuring real things. The interpretations require context, and that context is yours to build over time.

---

FAQs

What is the most important sensor in a biometric wearable?

It depends on your goal. For recovery and stress monitoring, HRV derived from the PPG sensor is arguably the most informative. For sleep quality, the combination of PPG and accelerometer data gives you the most actionable picture. For cardiovascular health monitoring, ECG is the most clinically relevant sensor.

How accurate is heart rate tracking on a wearable?

PPG-based heart rate tracking is generally accurate at rest and during steady-state aerobic activity. Accuracy drops during high-intensity interval training, weightlifting, and other activities with significant wrist movement. No consumer wearable matches the accuracy of a chest-strap ECG monitor during exercise.

Is HRV a reliable health metric?

HRV is a reliable relative metric when tracked consistently over time. Your own baseline trend is what's meaningful. Single-day readings are noisy and shouldn't be over-interpreted. Comparing your HRV to someone else's isn't useful, ranges vary significantly between individuals.

Can a wearable accurately detect sleep stages?

Consumer wearables estimate sleep stages using motion and heart rate data. They're useful for identifying patterns and trends in your sleep over time, but they're not as accurate as clinical polysomnography, which uses brain wave measurements alongside other physiological signals.

What does VO2 max estimation actually mean on a wearable?

It's a model-based estimate of your aerobic capacity derived from the relationship between your heart rate and your exercise output. The absolute number matters less than the trend. If your estimated VO2 max improves over a training period, that reflects real fitness progress, even if the number itself isn't clinically validated.

Why do different wearables give different readings for the same metric?

Different devices use different sensor hardware, sampling rates, and algorithms to process raw data. A heart rate reading from one device may differ from another because the signal processing and noise filtering aren't the same. This is why consistency matters more than absolute numbers, tracking one device over time gives you more useful data than comparing two devices against each other.

Do I need multiple sensors to get useful health data?

Not necessarily. The most useful everyday metrics (resting heart rate, HRV trends, sleep duration and staging, movement volume) can all be derived from PPG and accelerometer data alone. ECG, SpO2, and skin temperature add specific value for specific use cases, but they're not required for general health and recovery tracking.

---

Understanding what your wearable is actually doing changes how you use the data. The sensors are measuring real physiological signals. The question is whether the device you choose presents that data in a way that's genuinely useful, without requiring you to replace the watch you already wear. If that last part matters to you, it's worth seeing what hiyd.ai is building.