Menstrual cycle tracking has quietly become one of the most important features on a modern wearable. Every major device now includes a cycle module — Withings, Apple Watch, Oura, Fitbit, Garmin. The features look similar, the marketing language sounds similar, and the underlying accuracy, privacy exposure, and clinical usefulness are not similar at all. This is an honest walk-through for anyone who wants to track a cycle with a wearable, including the parts the app onboarding screens tend to skip.
The physiology the device is trying to detect
The menstrual cycle divides roughly into two phases separated by ovulation. The follicular phase begins on the first day of a period and lasts variably (often 10 to 20 days) until ovulation. Estrogen rises through this phase. The luteal phase begins immediately after ovulation and lasts a much more consistent 12 to 14 days until the next period. Progesterone dominates the luteal phase.
Two shifts around ovulation are the ones that consumer wearables can, in principle, detect.
- Basal body temperature rise. Progesterone is thermogenic. Once ovulation occurs, core temperature rises by roughly 0.2 to 0.5 degrees Celsius (about 0.4 to 0.9 degrees Fahrenheit) and stays elevated through the rest of the luteal phase, dropping again just before or at the start of menstruation. This is the biphasic pattern that the traditional fertility awareness method relies on.
- Overnight heart rate and HRV changes. Resting heart rate typically increases by 2 to 3 beats per minute in the luteal phase, and heart rate variability typically drops modestly compared to the follicular phase. These are real, measurable shifts that pattern across the cycle.
Modern wearables combine these signals. Most also integrate user-logged period start dates and adjust their predictions accordingly.
Natural Cycles and the FDA clearance that started this
In 2018 the FDA cleared Natural Cycles as a Class II medical device for use as a contraceptive. It was the first app in the US to receive that clearance. The system uses a traditional oral basal body thermometer taken each morning before getting out of bed, combined with the app's algorithm to identify fertile and non-fertile days.
The clearance came with published effectiveness data: with perfect use (following all the app's rules exactly), pregnancy rates were about 1 to 2 per 100 users per year. With typical use — real people, real inconsistencies — pregnancy rates were roughly 6 to 8 per 100 users per year. For comparison, a copper IUD is under 1 per 100 with typical use, hormonal birth control pills are around 7 per 100 with typical use, and no contraception at all is around 85 per 100.
Natural Cycles' clearance was a landmark. It also demonstrates the limit of temperature-based contraception. Even with a purpose-built oral thermometer and a validated algorithm, the typical-use failure rate is higher than most people expect.
The wrist and ring wearables in the market are not the Natural Cycles clearance. Some brands have partnered with Natural Cycles to feed wearable temperature data into the app — Apple Watch, Oura, and Garmin all have such integrations for parts of that stack — but the wearable itself is not FDA-cleared as a contraceptive. Read that sentence again if you are wearing one and planning cycle tracking around it.
Retrospective versus prospective detection
This is the distinction that decides whether a wearable is useful for a given cycle-tracking goal.
Retrospective detection means the device or app tells you, a day or two after the fact, that ovulation probably occurred on a particular day. It is looking backward for the sustained temperature shift. Every wrist and ring wearable in the market does this reasonably well when the signal is clean.
Prospective detection means the device tells you today that ovulation is going to occur within the next 24 to 48 hours. This is what matters for anyone trying to time conception or, more consequentially, trying to avoid it.
The honest truth: wearables are much better at retrospective detection than prospective detection. The temperature shift they rely on happens after ovulation, not before. HRV changes are similar. There is no clean physiological signal on a wrist that reliably says "ovulation tomorrow."
For prospective detection, the well-validated tools are luteinizing hormone (LH) urine test strips. LH surges in the roughly 24 to 36 hours before ovulation, and a positive LH strip is one of the most reliable early warnings you can get. LH strips are cheap (often under $1 each in bulk), accurate, and require no algorithm. If your goal is fertility timing, LH strips beat wearables, and it is not close.
The wearable's role is complementary. It confirms that ovulation happened, characterizes the luteal phase length, tracks the temperature curve across many months for pattern-finding, and integrates with sleep and stress metrics to give you a fuller picture of how your cycle interacts with the rest of your physiology.
Whose data trains the model, and where does it live?
Cycle data is among the most sensitive personal information a wearable collects. The relevant privacy questions have changed materially in the past few years.
HIPAA does not cover consumer wellness apps
A common misunderstanding: because cycle data is health data, it must be protected by HIPAA. That is not the case. HIPAA — the US Health Insurance Portability and Accountability Act — governs "covered entities," which means healthcare providers, health insurance plans, and healthcare clearinghouses, plus their business associates. A consumer wearable company selling you a watch is not a covered entity. A period tracking app you downloaded from the App Store is not a covered entity. HIPAA does not apply.
What actually governs consumer wellness apps is a patchwork of the FTC Health Breach Notification Rule (updated in 2024), state privacy laws (California's CCPA and Washington's My Health My Data Act being the most consumer-protective), and the app's own privacy policy. That is a much weaker floor than most people assume.
Flo, Ovia, and the enforcement history
In 2021 the FTC settled with Flo Health, alleging that Flo had shared user cycle data with Facebook, Google, and analytics providers despite promising users their data would be kept private. Flo did not admit wrongdoing but agreed to an audit. Ovia Health had a related settlement over data-sharing with employers as part of employer wellness programs. Both companies are still in operation. These histories are worth knowing before you decide whose app holds your cycle log.
The post-Dobbs US context
In 2022 the US Supreme Court decision in Dobbs v. Jackson Women's Health Organization ended federal constitutional protection for abortion access, and multiple states criminalized abortion under various conditions. Cycle-tracking data — showing missed periods, suspected pregnancy, and gaps consistent with terminated pregnancies — became legally interesting in ways it had not been before.
Since 2022, several period tracking apps have publicly committed to reducing data retention, moving to on-device storage, disabling third-party analytics for cycle data, or accepting anonymous accounts. Some have moved primary data storage out of the US. Enforcement of subpoenas against cycle tracking data has been limited but real, and prosecutors in some states have used data from digital sources — including search history and messaging apps — in cases related to pregnancy loss.
This is a genuine consideration for anyone in a state with restrictive laws. It is not paranoia. It is prudent operational security.
Where the data actually lives on a wearable
The answer differs sharply by manufacturer.
- Apple Watch cycle data is stored end-to-end encrypted with Advanced Data Protection turned on. Apple states it cannot access the data. This is currently the strongest privacy posture in the major-manufacturer market.
- Withings is a French company, and by default its servers are located in the European Union, which puts stored data under GDPR. Withings' privacy policy explicitly commits to not selling personal data. This is a stronger structural posture than most US-based competitors.
- Oura is a Finnish-headquartered company and follows GDPR for stored data.
- Fitbit is now owned by Google. Google has publicly committed to keeping Fitbit health data separate from advertising uses, but the parent-company context matters to some users. Google has stated it will delete abortion clinic visits from user location history, but the operational specifics are not fully public.
- Garmin stores data on its own servers and publishes what it does with it, but its detailed retention and law-enforcement disclosure practices are less documented.
The general rule: European-headquartered wearables benefit from GDPR by default. Apple's on-device encryption is the strongest single technical protection. Everyone else is closer to the median, which is not zero and not maximum.
Practical guidance for cycle tracking on a wearable
Here is what actually works.
- Use the wearable for pattern data, not for time-sensitive fertility decisions. Six months of nightly temperature deviation plotted against period start dates will show you whether your luteal phase is consistent, whether the temperature rise is clean, and whether you are having anovulatory cycles. That is real, useful information.
- Add LH strips for prospective timing. If you are trying to conceive or trying to avoid conception through fertility awareness, LH strips are the tool. Test in the mid-cycle window, twice daily if you have irregular cycles.
- Do not use a wearable as birth control. None of the devices in our catalog are FDA-cleared as contraceptives. Even the Natural Cycles clearance, which is the closest we get, has typical-use failure rates several times higher than long-acting reversible contraception.
- Read the privacy policy for the app you actually use. On an iPhone, Apple Health is currently the strongest storage option for cycle data. On Android and elsewhere, Withings and Oura are structurally cleaner than most alternatives. If you are in a state with restrictive laws, consider on-device-only apps like Euki or storing your cycle data manually rather than in a cloud service.
- Give the wearable time to learn your cycle. Two to three complete cycles of continuous wear are usually needed before temperature-based predictions stabilize. Short or highly irregular cycles are harder, and the algorithms perform worse on them.
Where our catalog stands
The Withings ScanWatch 2 is the wearable in our catalog best-suited to cycle tracking on the wrist. It measures continuous skin temperature, HRV, resting heart rate, and respiratory rate — the physiological signals that pattern across the cycle — and Withings' EU headquartering and published privacy policy give it a stronger structural privacy posture than most US-based alternatives. It is not FDA-cleared as a contraceptive.
The Fitbit Sense line has skin temperature, HRV, and mature cycle-tracking features integrated with Fitbit's app. It is a strong choice for someone already in the Fitbit ecosystem, with the caveat that the app now sits under Google.
None of the sleep-only devices, blood pressure monitors, or CGM sensors we carry meaningfully substitute for a cycle-tracking wearable. This is a wrist or ring category.
The honest bottom line
Wearable cycle tracking has become good enough to give you a useful longitudinal picture of your cycle without a daily oral thermometer. It has not become good enough to time ovulation precisely or to substitute for medical-grade contraception. It also lives in a privacy environment that is materially different from what most users assume, and materially different from what it was five years ago.
Use the wearable for the pattern. Use LH strips for the timing. Read the privacy policy of the app you actually use. And be honest with yourself about the difference between a health tracker and a medical device — none of the wrist and ring wearables in the current market are the second thing, even when the marketing slides suggest they might be.