Method
Everything the projection does, written out, so you can decide whether to believe it.
The lean fraction
The starting point is what share of lost weight came from lean tissue in the published body-composition data for each drug class.
- Tirzepatide (Zepbound, Mounjaro): about 25%, from SURMOUNT-1.
- Semaglutide (Ozempic, Wegovy): about 40 to 45%, from the STEP 1 body-composition substudy.
- Liraglutide and others: a wider band, because the published body-composition data is thinner. A 2026 review in Metabolites puts lean soft tissue at 26 to 40% of weight lost across recent trials of this drug class.
Trial participants were not following a resistance-training or high-protein protocol. That matters: the fraction above is the trajectory without either lever, which is why it is used as the baseline rather than as a prediction.
The two adjustments
Protein intake and resistance training both reduce lean-mass loss during weight loss. The direction of that effect is well established, and the 2026 clinical nutrition guidance names 1.2 to 1.6 g of protein per kg per day alongside progressive resistance training as the intervention. The size of it, for a person on a GLP-1 at a particular intake and training frequency, is not something any single trial gives a coefficient for.
So the adjustments here are conservative and directional, not measured. That is the reason the output is always a range with an explicit band, and never one confident number. If a product shows you a single precise figure for this, it is inventing precision it does not have.
What it is not
- It is not a measurement. A DEXA or InBody scan measures. This projects.
- It is not a warning on its own. A projection is a fixed function of what you typed, so it cannot notice that your body is doing something the trial average did not. That only shows up when a real reading sits beside it.
- It is not a diagnosis, and it is not medical advice.
- It will never recommend a dose, a change in dose, or whether to stay on the medication.
Where it is weakest
With only a weight and a goal this is an expectation, not a check. It says what the trial average does with the pounds you have left. It cannot say whether you are on that line, because it has nothing from your body to compare against. A body fat percentage from a smart scale gives it one. A DEXA gives it a better one, and the gap between expected and measured is the only part of this that is about you rather than about a trial population. Consumer impedance scales are noisy, so trends are shown rather than single readings, and the confidence state on every projection says how much the estimate actually knows about you.
The reading that argues against this
In May 2026 a group from the Medical University of Vienna presented work at the European Congress on Obesity on 486 adults treated with GLP-1 drugs. Fat mass fell by 9.0 kg, skeletal muscle mass by 1.2 kg, and relative skeletal muscle mass was preserved or increased in more than 70% of them. The headlines that followed said GLP-1 drugs preserve muscle. That study is on this page because anyone checking this product will find it, and because it deserves a straight answer rather than silence.
Three things separate it from the trials this projection uses. It measured with bioelectrical impedance rather than DEXA, and impedance infers tissue from body water, which shifts during rapid weight loss. It reported skeletal muscle mass, which is a smaller quantity than lean mass: lean mass also contains water, glycogen, connective tissue and organ tissue, so losing a quarter of your weight as lean mass and only a twentieth of your skeletal muscle are not contradictory findings. And it was retrospective with no control group, which its own authors state.
The honest position: the DEXA trials are the better instrument and this projection is built on them, but the question is genuinely open and the effect is probably smaller for someone eating enough protein and lifting than the trial averages suggest. That is the entire reason the two levers exist here.
Sources
- SURMOUNT-1 tirzepatide, body-composition substudy. Eli Lilly, registry entry with results.
- STEP 1 semaglutide, body-composition substudy. Novo Nordisk, registry entry with results.
- Medical nutrition in the GLP-1 era protein strategies, micronutrient monitoring and lean mass preservation, 2026.
- Lean Mass and Musculoskeletal Preservation in GLP-1-Based Obesity Treatment Metabolites, 2026.
- Medical University of Vienna, ECO2026 the real-world impedance study set out above, as reported.
Every number on the home page links to the entry it came from. If one of these is wrong, or if a better source exists, say so and it gets corrected or removed.