Why is women's cycle data so fragmented, and why does it matter?

Women have never recorded their cycles so much. The three largest menstrual apps have been downloaded more than 250 million times, and Flo alone reports around 70 million monthly active users. Add hormone monitors, wearables, cycle-observation methods and medical records: the feminine cycle is one of the most-documented phenomena in personal health.

And yet almost none of it can serve medical research. To understand why, you have to look closely at where this data lives, and in what state.

A lot of data, very little information

Take Flo, the largest app in the world. Most of what it collects comes down to one piece of information: the date of the last period. That is enough to predict the next one, but far too thin for any serious study. No observed biological sign, no hormone measurement, nothing that helps understand a cycle; only enough to date it. Multiplying 70 million users by such a thin data point does not produce science.

At the other end of the spectrum, an app like Read Your Body, valued by women who practise a cycle-observation method, collects rich, detailed observations. But everything stays stored on the phone. That is an excellent choice for privacy; it is also a dead end for research: the app offers no way to share data with a study, and no consent framework that would let a woman take part in one.

In between, each school of cycle observation describes the same biological sign, cervical fluid, in its own vocabulary. The Billings method records a sensation and an appearance; the Creighton Model uses numeric codes from 0 to 10; FEMM sorts observations into four visual categories. Rigorous records, sometimes kept for years, but in languages incompatible with one another.

And around those records: hormone measurements in a device maker’s cloud, lab results in one clinic’s system, an ultrasound in another’s. For care, the clinician sees one fragment at a time. For research, the longitudinal series that would answer the real questions (how symptoms, hormones and clinical outcomes relate across months and years) exists nowhere in a usable form.

Why the market won’t fix this

Each app publisher or device maker has an incentive to keep data inside its own product: interoperability is a cost, and data is a competitive asset. And the apps holding the largest volumes have systematic failures: a peer-reviewed study of the 23 most popular women’s health apps found 20 sharing data with third parties, without users being clearly aware of it. Data collected that way cannot anchor ethical research, however abundant it is.

The result is a paradox: the largest pool of women’s health data ever assembled is at once too thin, too locked up and too poorly consented to serve the women who produced it.

Our answer: an interoperability model for the feminine cycle

Health Data Safe has built what is, to our knowledge, the first open, publicly documented conversion layer between the scoring systems of the major cycle-observation methods. The principle: every category of every method (a Creighton “sticky”, a FEMM “slippery”, a Billings “Peak” day) is translated into nine common, weighted clinical dimensions: threadiness, stretchability, lubrication, transparency, wetness, fluidity, sensation, volume, colour.

Once that common language exists, what was impossible becomes simple: a single study can bring together participants who chart with Billings, Creighton, FEMM or a sympto-thermal method, and compare their observations on a shared scale. Quantitative hormone measurements (such as those from the Mira monitor) project onto the same scale. The model is open source: any researcher can examine it, criticise it and use it.

You can explore it yourself below.

One note of honesty: this model is operational, but its clinical reliability has not yet been independently confirmed. A peer-reviewed publication is in preparation, and the model is being submitted for evaluation by recognised experts in fertility-awareness medicine. Until then, it should be treated as an open research instrument, not as a validated diagnostic tool.

Defragmentation, in practice

The interoperability model is the common language; the Health Data Safe platform is the place where the fragments converge, under the person’s control. As of June 2026, this is deployed in production:

• Mira, the quantitative hormone monitor, is connected directly: hormone measurements join the person’s record.
• athenahealth, one of the major American electronic health record systems, is integrated: clinical data can follow the patient.
• Records from FEMM, Cyclefeminin.net and Read Your Body can be imported by file: observations locked inside a phone find a path back into the full record, if their owner decides so.
• A bridge to REDCap, the tool that university hospitals and research teams use to collect data for their studies, is being deployed: a researcher can receive a consented longitudinal series in a single operation.

Every connection goes through explicit consent: time-limited, tied to a precise purpose, revocable at any moment. The woman holds the complete picture of her cycle; everyone else sees exactly what she permits, no more, no less.

This is the foundation of our Women’s Health project. For the wider context: the women’s health data gap.