Nobody can tell you your number, but the spread is documented and it is wide. Across 135,349 people treated with semaglutide or tirzepatide, the share losing more than 15% ran from 10% to 34% depending on the brand[2]. Moderate response, between 5% and 15%, was remarkably stable at 40% to 42% across all four brands. The middle of the distribution is predictable; the top of it is not.
The brand differences are mostly dose and label rather than molecule. Ozempic and Mounjaro are diabetes-labeled, Wegovy and Zepbound the higher-dose obesity brands of the same two molecules, which is why a super-responder rate is partly an artifact of what was prescribed.
Early progress predicts later progress, though not as strictly as a seller’s cancellation window assumes. Pooling tirzepatide participants from two trials and splitting them at week 8 by whether they had lost 5%, both groups reached clinically meaningful reduction by week 72 [3]. Early responders did significantly better, but a slow start at eight weeks was not a failed course.
Predicting who responds would be commercially valuable to every seller on this desk, and clinically useful to everyone else. A study trained seven machine learning approaches on real-world records to try [1]. How much response actually varies is documented in the super-responder brand comparison.
The headline is a weight-loss model with discrimination around 0.94 and accuracy of 0.89 to 0.90, which in most clinical prediction contexts would be exceptional. The same methods on the same data predicted glycemic control at around 0.79 discrimination and 0.73 accuracy — respectable, and far short of the first figure. Expressed as plain accuracy that is 89% vs 73% of predictions correct.
The gap between the two models is the more informative result. Glycemic control is the harder target and the less self-referential one, and it is where the honest ceiling of this approach shows. Roughly three predictions in four came out correct, built mainly on duration of diabetes, baseline HbA1c and existing sulfonylurea or insulin use. That is a useful tool and not a crystal ball. It is also a more honest number than the weight model's, for the same reason the derived figures in the prediabetes screening analysis need their predictive values read rather than their sensitivities.
One methodological limit matters commercially. The models were assessed with ten-fold cross-validation, which resamples the same dataset, and no external validation on a separate population is described. A model can score highly on the data it grew up in and fail on patients from a different health system. That is precisely what would happen if a seller took a published model and ran it on their own customers.
There is also nothing here a buyer can act on. The predictors are baseline characteristics, so learning that your starting BMI predicts your response does not give you anything to change. The measurements that do shift with treatment are handled in responding is not one thing, and the question of how early a real response becomes visible in slow responders at week eight.