The prescription arrived. You've been waiting weeks — maybe months. You've done the research, had the conversation with your GP or specialist, and now the box is sitting on your kitchen counter. Ozempic. Wegovy. Semaglutide.
Now what?
For most people starting semaglutide, the plan is: inject, wait, watch the scale. And when the weight moves — or doesn't — they conclude that it's working or it isn't. The problem, if you have PCOS or PMOS, is that the scale is the wrong instrument entirely. Semaglutide can be reducing your insulin resistance, quietening the food noise your hypothalamus has been generating for years, and gradually shifting your androgen balance — all while the scale moves slowly, erratically, or in the wrong direction for weeks.
Research confirms that semaglutide reduces insulin resistance and free androgen levels in PCOS — but the effect varies significantly depending on your baseline insulin sensitivity and cortisol patterns. Two women with the same diagnosis, same starting weight, and same dose can have completely different responses at 8 weeks based on the underlying mechanisms driving their symptoms.
Ozempic isn't a PCOS cure. It's a metabolic intervention. Tracking is what tells you whether the intervention is matching your specific pattern.
What Semaglutide Actually Does in a PCOS/PMOS Body
Semaglutide is a GLP-1 receptor agonist — it mimics glucagon-like peptide-1, a hormone your gut releases after eating. In a PCOS/PMOS body, this matters for four connected reasons.
It slows gastric emptying and blunts the post-meal glucose spike
When food moves more slowly out of the stomach, glucose enters the bloodstream more gradually. The sharp post-meal insulin spike — the one that's exaggerated in insulin-resistant bodies — becomes blunted. Over time, a lower insulin pulse means the ovaries receive less constant insulin signalling. In PCOS/PMOS, insulin excess drives the ovaries to overproduce testosterone. As insulin drops, androgen production gradually reduces with it. This is not a fast mechanism — it unfolds over weeks and months — but it's one of the core pathways through which semaglutide improves PCOS/PMOS metabolic markers.
It activates GLP-1 receptors in the hypothalamus
The hypothalamus has its own GLP-1 receptors. When semaglutide activates them, it reduces appetite and — crucially for PCOS/PMOS — quietens food noise. That persistent, intrusive background chatter about food: what you'll eat, whether you should, what you had earlier, what you'll have tomorrow. In PCOS/PMOS, food noise is partly hypothalamic and partly driven by reactive glucose patterns. The hypothalamic effect is often the first change users notice — and it typically arrives within 2–4 weeks.
Weight loss reduces visceral fat and shifts oestrogen balance
As weight reduces, visceral fat (the metabolically active fat stored around the organs) decreases. Visceral fat contains an enzyme called aromatase that converts androgens into oestrogen. In PCOS/PMOS, where oestrogen balance is already disrupted, less visceral fat means less aberrant aromatase activity — and gradually, improved oestrogen/androgen ratio. This is a downstream effect, slower to show than the appetite and insulin changes.
How it differs from Mounjaro — and why it matters
This is worth saying clearly, because the confusion between the two medications is widespread. Mounjaro (tirzepatide) is a dual agonist — it activates both GLP-1 and GIP receptors. The GIP component adds a more direct effect on body composition, favouring muscle retention and more pronounced insulin sensitisation, and the dual mechanism tends to produce greater weight loss on average. Semaglutide is a single GLP-1 agonist — no GIP activity. The body composition split (muscle vs fat) is less favourable than with tirzepatide, and insulin sensitisation, while real, is less pronounced. If you're comparing options or have already tried Mounjaro, this post on Mounjaro and PCOS/PMOS weight loss covers the dual-agonist angle in detail. The point here isn't that semaglutide is less effective — it's that the mechanisms are different, and so the signals to track are different.
What Your Atlas Check-in Data Is Telling You
If you're tracking daily in Atlas after starting semaglutide, five signals will start to tell you the story before the scale does.
Food noise and evening craving score
The hypothalamic effect is usually the first to arrive. Most people notice it as a reduction in evening food noise — that late-night rummaging impulse, the 9pm "I need something" feeling that's partly appetite and partly habit and mostly hypothalamic. In your Atlas check-ins, watch your evening craving score. A consistent downward shift over 2–4 weeks — even a partial one — is your first indication that semaglutide is engaging the hypothalamic pathway.
For a deeper look at what's driving food noise in PCOS/PMOS bodies, this post on the night sugar craving pattern covers the mechanism in full.
Post-meal energy at 90 minutes
Reactive hypoglycaemia — the crash that hits roughly an hour after eating — is common in insulin-resistant PCOS/PMOS bodies. You eat, insulin spikes too high, blood glucose drops too far, and you're tired and craving sugar again before the meal has fully digested. Semaglutide's gastric emptying effect should start to blunt this. In your Atlas check-ins, look at your energy rating at approximately 90 minutes post-meal. Episodes of sharp post-meal energy drops should become less frequent. This signal tends to appear in weeks 3–6.
Cycle regularity and length
Androgen reduction is a slow process. Expect 8–12 weeks minimum before any cycle-related shift becomes visible. What to watch: cycle length moving toward 28–35 days; the gap between cycles becoming more consistent; the pre-menstrual symptom cluster (bloating, mood, cravings) becoming marginally less intense. Don't assess this signal at 6 weeks — you'll draw the wrong conclusion.
Morning energy baseline
One of the indicators of improving insulin resistance is improved morning energy — the sense of waking up with some capacity rather than already depleted. This signal typically emerges from week 8 onward as the insulin pathway improves. The important caveat: weeks 1–4 often bring nausea, particularly as the dose is titrated upward. Morning energy will likely be suppressed during this phase. Log it accurately, but don't interpret early nausea-related energy dips as evidence the drug isn't working. They're not — they're a titration effect, not a metabolic one.
Skin and acne tag
Free androgen reduction shows in skin, but it's the slowest signal by far. Expect 10–16 weeks from the point androgen levels drop before skin changes become visible. Androgen-driven acne — particularly chin and jaw clusters — is one of the last things to shift, not the first. If you're logging skin flares in Atlas, you're building the baseline now that will tell you whether the pattern is improving in month 4 or 5. The data point isn't the current flare; it's the trajectory.
Three Patterns Worth Tracking Over 8–12 Weeks
Food noise trend: week 1 vs week 4 vs week 8
Log your evening craving score consistently — ideally at the same time each evening, before eating anything. Compare the average of your first full week with week 4, then week 8. The flattening of this score over time is the clearest early signal that semaglutide is working on the hypothalamic pathway. An improving score doesn't mean the cravings disappear — it means they become easier to sit with, less insistent, less consuming. That's the mechanism working.
Post-meal glucose proxy: energy at T+90min across meal types
This one requires a bit of deliberate variation. Over an 8-week period, try to log your T+90min energy across a range of meal types: protein-first meals, carb-heavy meals, mixed meals. As semaglutide continues, the gap between the energy scores for different meal types should narrow. A high-carb meal that used to guarantee a 2pm slump should begin to produce a smaller crash as gastric emptying slows and the glucose spike blunts. Narrowing of the meal-type gap is a proxy for improved glucose regulation.
> "Most people assess Ozempic by the number on the scale. But in a PCOS/PMOS body, the number that matters first is your evening craving score."
Cycle length trajectory: months 2, 3, and 4
Discount the first cycle after starting semaglutide. Nausea, reduced food intake, and the initial metabolic disruption make it unreliable. From month 2 onward, plot three consecutive cycle lengths. A trajectory moving toward 28–35 days — even incrementally — is the androgen-reduction signal arriving in the data. Atlas's premium pattern engine overlays your cycle length history against your food noise trend so you can see whether the hormonal shift is tracking the metabolic one. These two signals should converge: as food noise reduces and insulin improves, cycle length should follow — on a lag of roughly 4–8 weeks.
What to Do With the Insight
Establish a baseline before you start (or as soon as possible)
If you haven't started semaglutide yet, use the next 4 weeks to build a baseline in Atlas: food noise, cycle day, post-meal energy, morning energy, skin tag. If you've already started, begin now — even a retrospective 2-week baseline is better than none. The baseline is what makes the trend legible. Without it, "things feel better" is anecdotal. With it, "my evening craving score has dropped from 7.2 to 4.1 over six weeks" is data.
Assess at the right intervals
The instinct is to check everything at once. Resist it. 4 weeks: assess food noise and appetite signal only. 8 weeks: add energy and post-meal response. 12 weeks: add cycle length trend. 16 weeks: add skin and androgen signal. Each signal operates on its own timeline. Assessing cycle change at 4 weeks will tell you nothing, and interpreting silence as failure will lead you to the wrong conclusion.
Frame the GP conversation around trends, not weight
The standard GP check-in for semaglutide is almost entirely weight-focused. That's the wrong lens for a PCOS/PMOS body. Bring your Atlas trend data: "My evening cravings have reduced from an average of 7 to 3 over six weeks" and "my last cycle was 38 days versus the previous two at 45 days" is more clinically meaningful than your current weight reading. This post on tracking metformin for PCOS/PMOS covers the GP framing approach in detail — the same logic applies here.
If your food noise signal has responded clearly but your cycle hasn't shifted at 12 weeks, that's a specific signal. Bring it. It suggests the hypothalamic pathway is responding but the androgen pathway is lagging — which may point toward a need for an expanded hormonal panel: free testosterone, SHBG, and a HOMA-IR re-check to confirm the insulin sensitivity picture. The data gives your GP something to act on rather than simply observe.
Comparing semaglutide with other options?
If you're still deciding between semaglutide and tirzepatide, or wondering whether the dual-agonist mechanism of Mounjaro would be a better match for your pattern, the Mounjaro PCOS/PMOS post covers that comparison directly. And if you're still questioning why the scale isn't the right measure for any of this, this post on why the calories-in/out model fails PCOS/PMOS bodies explains the underlying metabolic reasons.
Semaglutide is a serious metabolic intervention. For PCOS/PMOS bodies specifically, it works through mechanisms that the scale won't reflect for weeks or months. The women who get the most out of it are not the ones who feel the most or weigh the least — they're the ones who understand what's happening, why it's happening, and what signals to watch for.
Your evening craving score is the number to watch first. Start tracking it today.