The month-four plateau on a GLP-1: why it happens, and what to do about it

The 30-second summary

• A flatline in months three to six is the rule, not the exception. The biology has a name: metabolic adaptation.
• Three levers move it: protein and resistance training (protect the engine), dose review (check the climb), and sleep (the unsexy multiplier).
• The scale is not the only measurement. Waist, weekly average, and how clothes fit will move while the morning weight does not.

Why the body stalls

In the first eight weeks on a GLP-1, two things drive the scale down: a real fat-mass loss and a real water-mass loss. The second is fast. The first is slower.

By month three, the water adjustment is largely done. Now the loss is mostly fat. Fat loss is constrained by your daily calorie deficit, and your deficit is constrained by your daily energy expenditure, which has fallen, because you are smaller.

This is called metabolic adaptation. The math is unforgiving. A woman who needed 2,200 calories a day to maintain at 80 kg needs perhaps 1,900 to maintain at 70 kg. If she was losing weight at 1,500 calories a day in month one (a 700-calorie deficit), she is now losing at 1,500 calories with only a 400-calorie deficit. The pace slows. Sometimes it stops entirely for two or three weeks at a time. (Hall et al., Lancet Diabetes Endocrinol, 2016.)

This is not the drug failing. The drug is still working. The drug has done what it could do at the current dose; now the body has caught up.

What a plateau is, and isn't

A plateau is a flat weekly average over three to four weeks. One bad week is not a plateau. A four-day spike is not a plateau. Water weight can move two to three kilograms in either direction in a single week from sodium, hormones, sleep, training, or travel.

Look at the four-week moving average. If that line has not moved at all in 21 to 28 days, you have a real plateau. If it is just bumpy, you are still progressing, bumps are part of the line.

The first lever: protect the engine

The fastest way to make a plateau permanent is to lose so much muscle that your maintenance calories drop further. Less muscle, less daily burn, more adaptation.

The fix is the same as it was in month one, only now it actually matters:

• 120 grams of protein a day, distributed across meals. Most women on a GLP-1 are eating 60–80 grams without thinking. The difference between 70 and 120 is not subtle in terms of muscle retention. (Our protein article is the deeper dive.)
• Resistance training twice a week. Bodyweight squats, push-ups, rows. Twenty minutes. The point is the stimulus, not the intensity.

If you do nothing else differently at a plateau, do this. The plateau will not break the same week. But over four to six weeks, the curve usually resumes, at a slower pace, but a real one.

The second lever: dose review

Most plateaus are not a dose problem. But some are. If you have been on the same dose for more than 12 weeks and the four-week average has not moved at all, this is a conversation worth having with your prescriber.

There are three possible answers from a prescriber:

1. Move up a dose. The standard titration schedule has a max: for semaglutide, 2.4 mg weekly; for tirzepatide, 15 mg weekly. If you have not reached it and the plateau is real, the next step on the schedule may unlock more loss.
2. Stay where you are. Sometimes the right answer is patience plus the first lever. If you have been losing weight too fast and your face/skin/energy are paying for it, the plateau may be the body recovering. A pause is not a failure.
3. Switch the drug. If you have plateaued at max dose of one drug and total loss is still meaningfully short of your goal, switching from semaglutide to tirzepatide (or vice versa) is a recognised option. The data from the SURMOUNT-5 head-to-head suggests tirzepatide produces about 50% more average loss than semaglutide. (Aronne et al., NEJM 2025.)

None of these is a decision for the internet. All three are reasonable answers from a prescriber depending on your context.

The third lever: sleep

The unsexy answer that nothing on the internet wants to tell you about.

Short sleep, under six hours, sustained for weeks, increases ghrelin (hunger), decreases leptin (fullness), worsens insulin sensitivity, and reliably stalls weight loss. (St-Onge et al., Am J Clin Nutr 2011.)

If your plateau coincides with a stretch of bad sleep, work stress, a sick child, a flight, a new schedule, sleep is the variable to fix first. The plateau often resolves within a week or two of consistent seven-hour nights, with no other change to diet or training.

What to measure instead of the scale

The morning scale is the noisiest measurement you take. The signal you want is in three other places:

1. The four-week moving average. Less noise. Calmer trend.
2. Your waist measurement. Once a week, in the morning, same level on your stomach. Visceral fat can change even when scale weight does not.
3. The fit of one piece of clothing. A pair of jeans, a fitted shirt. Honest about what your body is doing under the surface.

What Steady does with this

Steady's progress view shows the four-week moving average as the headline number, not the morning weight. Symptoms, nausea, fatigue, mood, are plotted next to the weight line, so a plateau plus a bad-sleep pattern is visible at a glance. When you sit down with your prescriber to talk about whether to titrate, you have the evidence, not the impression.

Sources

1. Hall KD, Kahan S. Maintenance of lost weight and long-term management of obesity. Lancet Diabetes Endocrinol 2018. PubMed
2. Aronne LJ et al. Tirzepatide vs Semaglutide for Obesity (SURMOUNT-5). NEJM 2025. NEJM
3. St-Onge MP et al. Short sleep duration and dietary intake. Am J Clin Nutr 2011;94:410-6. PubMed
4. Cermak NM et al. Protein supplementation augments lean-mass retention. AJCN 2012. PubMed

Medical disclaimer: Plateau-breaking strategies should be discussed with your prescriber, not the internet. See our full medical disclaimer.