The Gender Data Gap in Wearable IoT Is Bigger Than a Missing Period Tracker

Every time a new health wearable is released, there is a press release about “groundbreaking insights into your body”. Cool. But most of this breakthrough science comes from one type of body that wasn’t a woman.

This is a common complaint in the medical industry, but something worth pulling into this conversation, specifically regarding wearable IoT, as we tend to believe that a smart ring or a smart watch is somehow gender neutral. That’s not the case. 

From the sensors to the software algorithm, even from the design to how it fits on your wrist or finger, everything was developed through research that has largely ignored women. And this didn’t change because the device shrunk and the interface became prettier.

How We Got Here

To know how your smartwatch is working differently for you, we have to go back further than the device itself.

In 1977, the FDA issued guidance that essentially excluded all women of childbearing age from early-phase clinical drug trials. It was not a small or narrow rule either. It applied to women using contraception, single women, and women whose husbands had had a vasectomy. 

The intention was to protect against another drug tragedy like the thalidomide situation, but the result was that for 16 years, most new drugs and medical research were built almost entirely on male physiology.

The guidance was overturned in 1993, when the government itself admitted that the policy was too rigid and paternalistic. At the same time, the NIH Revitalization Act mandated inclusion of women in adequate numbers in NIH-funded research. 

So either way, modern medicine has had somewhere around three decades of intentionally researching the female bodies after about a century of mostly studying male ones.

But what does any of this have to do with a company trying to build a fitness band in 2026? Well, for starters, the baseline assumptions, the normal heart rate values, the sleep architecture profiles, and the resting metabolic equations that many wearable devices use were based on research that was biased against women. The discrepancy wasn’t created by wearable companies. They inherited it and then built products on top of it.

Where the Gap Actually Shows Up in Your Device

Here’s where we get into things that are easily overlooked since they sound theoretical, but aren’t. They manifest very concretely.

The Sensor

Fitness trackers and smartwatches usually employ photoplethysmography, or PPG, in measuring your heart rate. A light goes into your skin, your blood absorbs some of it, and the sensor measures the resulting reflections in order to calculate your heart rate. It’s a neat trick and usually works well for resting heart rate.

The issue emerges when you actually start using your device, especially when taking into account that there is a problem of body shape, which doesn’t seem to get much attention. Most wrist devices have been calibrated based on male physiology. 

The fact that people have smaller wrists, different fat distribution, and that a chest strap or an armband will feel different on a woman’s body makes a huge difference in terms of the sensor accuracy.

And when you factor in movement such as writing or cooking or holding a baby, you know how unreliable those results can get from certain activities. Never mind your body shape, which was not even properly tested in the first place.

The Hormone Blind Spot

Well, this is a big one. Many aspects of wearable health metrics- resting heart rate, heart rate variability, body temperature, sleep quality- all change during the menstrual cycle. In addition to affecting mood swings, estrogen and progesterone affect basal body temperature, HRV, and exercise response.

A cyclic baseline was not taken into account in most consumer wearables. The algorithms are calibrated to one static normal, and it works fine as long as your hormones are not changing much from month to month; it fails miserably otherwise. 

A gadget that indicates that something is abnormal without taking into account where you are in your cycle isn’t doing anything wrong. It just wasn’t designed to see that pattern in the first place.

Cycle tracking on its own has become a standard feature of wearables relatively recently. The Oura Ring partnered with Natural Cycles for ovulation tracking in August 2022 and launched their dedicated, proprietary AI model, Oura Advisor, for cycle tracking and women’s health in April 2026. 

It is one of the more cycle-forward brands out there. A lot of mainstream wearables added menstrual tracking as a bolt-on feature years after launch, almost as an afterthought, not as something core to how the device reads your body.

Even the Skin Sensor Issue Has a Gender Layer

The issue of skin tone bias in PPG sensors has been increasingly talked about recently. It is not a gender issue, but an issue nonetheless. Studies have demonstrated the possibility of receiving inaccurate measurements of PPG heart rate, 15 percent higher in dark skin than in light skin. The same goes for tattooed skin.

This is a different type of blind spot, and a darker-skinned woman is potentially dealing with two layers of sensor bias at once, not one.

FemTech Is Trying to Close the Gap, Slowly

The bright side of this is that the industry seems to be aware of the problem. Femtech as a category basically exists because of this exact gap. Clinical research for decades has been centered around male participants until the 90s and created a massive gap in data about female physiology and disease patterns, and Femtech aims to close this gap by generating a lot of data specific to female physiology.

This is reflected in the rapid rise in popularity of menstrual and fertility tracking apps to become an integral function of wearable devices. It is also reflected in products like the Vivoo Flow Pad and the continuous hormone tracker, Clair, coming into the market.

Regulators are also fostering some change. As recently as June 2024, the FDA released its draft guidance for diversity action plans relating to clinical trials for digital health technologies, which is certainly a good move because it ensures that the devices are tested by the people using them, not just a convenient sample size.

What This Really Means for People Designing These Devices

While designing wearable IoT devices, the takeaway from this is not simply, "Add a period tracking feature and you've done enough." There's much more to think about:

• Skin tone diversity must be taken into consideration when validating your product. You also need diverse anatomy and physiological conditions. Menopause can affect the body in the same way as puberty, but none of the wearables consider the different physiology of adult humans.
• Incomplete training of the algorithm means that it can hardly become complete afterward. If your models for heart rate variability and sleep were created based on data collected mostly for men, adding a “female mode” is not enough.
• The design of hardware is equally important. The location of the sensor, the fit of the band, and the way your smart ring is sized can affect what readings you get from whom.

All these things are not just trends to follow. Women account for about half of the people purchasing smart wearable devices and a big part of this growing market segment in women’s health applications.

Closing Thought

There is no single company to blame for the gender data gap issue in wearables. It was formed as a result of decades of research that neglected the female population and thus manifested through the sensors in your wearable and algorithms that determine your sleep score.

The companies that actually work towards solving the problem treat it as a design problem from the very beginning rather than an afterthought when something goes wrong.

We’re a few years into Femtech as a category instead of a niche. But years against decades is quite a long gap to close, especially when every wearable tells you that they know your body better than you do.