You've been told to track your ovulation. So you buy the ovulation predictor kits, you follow the app's fertile window, you do everything the fertility advice says. And then — nothing. The OPKs keep showing positive, or positive for days at a stretch, and you can't tell whether that's ovulation or noise. The app tells you your fertile window is day 12 to day 17. Your cycle is 41 days. None of it fits.
If you have PCOS or PMOS and you're trying to conceive, standard fertility tracking tools were not built for your body. Understanding why — and knowing what to do instead — is the difference between months of confusion and a data picture that actually tells you something useful.
Why PCOS/PMOS Makes Fertility So Confusing
The two most common fertility tracking methods — ovulation predictor kits (OPKs) and calendar-based fertile windows — both fail for PCOS/PMOS bodies in specific, predictable ways.
The OPK problem. OPKs work by detecting the LH surge that precedes ovulation. In a typical cycle, LH spikes sharply once, ovulation follows 12–36 hours later, and the OPK result is unambiguous. In a PCOS/PMOS body, baseline LH is often chronically elevated — meaning OPKs can read positive for days or even weeks without ovulation actually occurring. You're not detecting a surge. You're detecting a baseline that happens to look like a surge to a test designed for a different hormonal environment. Many women with PCOS/PMOS report weeks of positive OPKs with no confirmed ovulation, which is both confusing and deeply demoralising if you don't understand what you're seeing.
The calendar problem. Calendar-based fertile windows — including the estimates built into most cycle apps — are calculated by assuming that ovulation happens around day 14 of a 28-day cycle. If your cycle varies between 24 and 58 days across six months (which is entirely common in PCOS/PMOS), that calculation is meaningless. You cannot predict a fertile window from a cycle length that isn't consistent, and applying a day-14 rule to a 45-day cycle doesn't just miss ovulation — it often places the "window" in the follicular phase when ovulation hasn't happened yet.
The deeper issue is that anovulatory cycles — cycles where no ovulation occurs at all — are common in PCOS/PMOS. It's possible to have a period and bleed without ovulating; the bleed is a withdrawal bleed from oestrogen, not evidence of a functional cycle. You can have months of apparently normal periods while never actually ovulating, and without more granular tracking, this can go completely unnoticed.
What Your Tracking Data Actually Reveals
Here's the shift in thinking that matters: you may not be able to confirm ovulation in any given cycle. But you absolutely can identify patterns across three to six cycles — and those patterns are more useful for understanding your fertility than a single OPK result ever will be.
Cycle length variability over time. Not just this month's length, but the mean and variance across your last six cycles. Is your cycle gradually shortening, which might suggest improving insulin regulation? Is it wildly variable, which tells you ovulation is inconsistent? A mean cycle length of 38 days with ±5 days variability tells a completely different fertility story than a mean of 38 days with ±22 days variability. Both average out to roughly the same thing; only the longitudinal picture shows the difference.
Energy and mood across cycle phases. If you're tracking daily energy and mood, something interesting starts to emerge over several cycles: most women with PCOS/PMOS can identify a rough follicular window — a period of days where energy is higher, mood is brighter, libido may be elevated. Even without a confirmed LH peak, this energetic shift corresponds to follicular activity. It's not a precise ovulation detector, but it's a real signal. The pattern of energy crashes before your period has a mirror image: the days before the crash, where things feel relatively good, are worth paying attention to.
Post-ovulation symptom proxies. Without temperature tracking or confirmed progesterone testing, your symptom data can still give you a progesterone proxy. Fatigue that arrives in a consistent window — say, around day 18–22 regardless of cycle length — is a potential luteal-phase signal. Breast tenderness in the second half of your cycle. Cravings that shift in character, from carbohydrate-driven food noise to something more specific and food-focused. These aren't confirmations of ovulation, but they're signals that something hormonal is happening in the second half of your cycle, which is more than an OPK in a high-baseline-LH environment will tell you.
Cravings correlated with cycle phase. Women with PCOS/PMOS often notice that their craving pattern shifts in predictable ways across the cycle — if they're tracking closely enough to see it. Intense carbohydrate cravings in what would be the luteal phase, specifically, can be a proxy for progesterone-driven insulin sensitivity changes. When you see that pattern repeat across three cycles, you've got a data point that's genuinely informative about whether your luteal phase is functioning at all.
The key argument here is simple: you can't reliably spot ovulation in a single PCOS/PMOS cycle using standard tools. But you CAN spot patterns across 3–6 cycles that tell you whether ovulation is happening at all, roughly when it's occurring, and whether your hormonal environment supports a viable luteal phase. That requires time and consistent data — not better OPKs.
The 5 Data Points Fertility Specialists Wish You Brought
Most fertility consultations involve a handful of blood tests, an ultrasound, and a BMI calculation. What they don't involve is any inquiry about your daily patterns — because clinics haven't had a way to collect that data, and appointments are too short to explore it. But here's what would actually be useful.
1. Cycle length over the last 6 months — mean and variance.
Not just "my cycles are irregular." The actual numbers: your last six cycle lengths, your average, and the spread. A mean of 36 days ± 8 days tells a specialist that you're likely ovulating, sometimes, inconsistently. A mean of 55 days ± 25 days tells them something very different. This is the foundational piece of data that contextualises everything else.
2. Your energy pattern by cycle day.
Which days in an average cycle do you feel your best? If you can identify a window — say, days 10–16 — where energy is consistently higher than baseline, that window is likely follicular. Knowing this gives both you and your specialist a reference point for approximate ovulation timing, even without confirmed LH data.
3. Mid-cycle symptom spike.
Even without a confirmed LH peak, do you notice a consistent mid-cycle moment — a day or two of heightened mood, a change in cervical mucus, a brief energy peak that then starts to soften? Log this when it happens. Over several cycles, even approximate, these markers can help identify whether ovulation is occurring and roughly when.
4. Post-ovulation cravings and fatigue onset.
The proxy for your progesterone response: does your craving pattern shift in the second half of your cycle? Does fatigue arrive in a predictable window after what you estimate as ovulation? Progesterone should rise sharply after ovulation and stay elevated through the luteal phase — but in many PCOS/PMOS bodies, it rises inadequately or too briefly. Your craving and fatigue patterns are a rough real-world proxy for whether that's happening.
5. Sleep disruption in the second half of your cycle.
Luteal insomnia is a documented PCOS/PMOS pattern that is almost never discussed in standard fertility contexts. Progesterone affects sleep architecture — but in PCOS/PMOS, the relationship between progesterone and sleep is disrupted. Women who struggle specifically with sleep in the 10–14 days before their period often have a cortisol and progesterone dysregulation pattern that's relevant to fertility. Tracking your sleep quality by cycle phase — specifically flagging whether it consistently worsens in the second half — is a genuine clinical signal. The cortisol and stress data post covers how luteal cortisol spiking shows up in tracking data and what it tells you about your hormonal environment.
Before IVF: The Data Conversation Your Clinic Won't Have
UK fertility clinics will typically see you for 20 minutes. They'll check your AMH (anti-Müllerian hormone, a marker of ovarian reserve), your AFC (antral follicle count on ultrasound), your BMI, and possibly your partner's sperm analysis. These are the numbers that determine eligibility and protocol. They are not the whole picture.
What a clinic won't ask about: whether your insulin resistance is currently active. Whether your cortisol spikes in your luteal phase. Whether your energy patterns suggest your follicular phase is functioning. Whether your cravings shift in ways that suggest a progesterone response. These things matter for IVF outcomes — specifically for how your body responds to ovarian stimulation, how it handles the progesterone supplementation given during an embryo transfer cycle, and whether your uterine environment is hormonally receptive.
None of that information comes from AMH and AFC. It comes from the kind of longitudinal daily data that most patients have never collected, because nobody told them to.
Atlas data functions as a second layer of evidence — something you bring alongside your clinic's blood test results to give a fuller picture of your metabolic and hormonal environment. If you've already had the data conversation with your GP, you know how much a well-organised pattern summary can shift a clinical conversation. The same applies to a fertility consultation. Your energy patterns, craving data, sleep quality by cycle phase, and longitudinal cycle variability all represent information your clinic genuinely doesn't have — and that, in some cases, will be directly relevant to your protocol.
If your PCOS/PMOS is driving insulin resistance that's still active, that has implications for how your body will respond to gonadotropin stimulation and the risk of ovarian hyperstimulation syndrome (OHSS). If your luteal cortisol is spiking, that has implications for your endometrial environment during a transfer. These are questions worth raising — and tracked data is how you raise them credibly.
What to Track Starting Today (Even If IVF Is Years Away)
Pattern clarity requires a minimum of three months of data. Ideally six. This is not something you can compress: the patterns only become visible when the same signals have repeated across multiple cycles, in multiple hormonal contexts. If IVF is two years away, you have two years to build one of the most detailed personal fertility datasets your clinic will have ever seen. If IVF is next year, start now — because every month of data you have is a month of evidence you wouldn't otherwise.
The daily check-in takes 30 seconds. Cycle day, energy (1–5), cravings (1–5), mood (1–5). That's the core. Add food tagging for insulin response — specifically flagging high-carb meals and noting how you feel 90 minutes later — and you're building a picture of whether insulin resistance is active and how significantly it's affecting your daily patterns. What 90 days of data reveals goes into detail on how the pattern engine surfaces these correlations automatically once there's enough signal to work with.
You don't need to track perfectly. Missed days don't reset the data. Even imperfect, inconsistent tracking over three to six cycles produces a longitudinal picture that no single test and no single appointment can provide.
Start now. The fertility picture you'll walk into a clinic with in six months — or six years — is built one daily check-in at a time.
Atlas starts building your fertility picture from day one — see Atlas plans. Your cycle data, your energy patterns, your craving shifts — all building quietly in the background, month after month, into the kind of evidence that actually moves conversations forward.