Most goal-weight advice fails in the same direction: it takes a number from a trial, subtracts it from a starting weight, and calls the result a goal. The numbers deserve better handling than that. A trial average is the center of a wide spread, computed on the people who stayed in the trial, measured at a fixed endpoint that took well over a year to reach. None of those three facts makes the number useless. All three change what it is fair to do with it.
This page is about choosing a goal weight in a way that survives contact with real data. Its subject is not the arc of results, which is covered at Month by Month: What GLP-1 Results Actually Look Like, nor life after the target, which has its own page at Maintenance Mode: Tracking After You Reach Goal Weight. It is the narrower question of where a defensible target comes from in the first place.
What a trial number actually is
Start with how the headline figures were earned. In the Zepbound weight reduction studies, weight was assessed after 72 weeks of treatment, at least 52 weeks of which were at the maintenance dose, and the mean baseline body weight in the first study was 100.7 kg with a mean BMI of 36.1. The Wegovy weight trials ran to Week 68. In a trial of adults with type 2 diabetes, the percent change in body weight was -9.4 percent on Wegovy against -0.9 percent on placebo, a difference from placebo of -8.5 percent. In its sleep apnea studies, Zepbound reported percent changes in body weight of -17.7 and -19.6 percent against placebo changes of -1.6 and -2.3 percent, differences from placebo of -16.1 and -17.3 percent.
Read that list once and note what it contains: populations with defined starting weights, fixed durations, and a placebo subtraction. Somebody who reads about 20 percent and multiplies it by their own starting weight has done the one piece of arithmetic the label never does. These are group averages over sixty-eight to seventy-two weeks in studied populations, and the label says so in its own way by publishing what it publishes alongside them.
The spread behind the average
That last sentence is the spine of this article. The primary efficacy parameters in the Zepbound trials were two things at once: the mean percent change in body weight, and the percentage of patients achieving at least 5 percent weight reduction. The label adds that among patients on the 10 mg and 15 mg doses, greater proportions of patients achieved at least 10, 15 and 20 percent weight reduction compared to placebo.
Ask why a regulator would require the second measure, and the answer is the point: an average describes nobody. If everyone in a trial lost exactly the mean amount, the responder thresholds would be redundant. They exist because the spread is wide, with substantial fractions of participants above and below the mean, and someone reading about a 20 percent average should understand that substantial numbers of participants did better than that and substantial numbers did worse. The Zepbound label also notes that a reduction in body weight was observed irrespective of age, sex, race, ethnicity, baseline BMI and glycemic status, which is a real statement about how broadly the effect appeared, and still a statement about groups.
GLP 1 Tracker AppWeight logged day by day, with progress photos and Apple Health import, kept privately on your phone. A month of readings beats one morning's number.
Get the app
The people who are not in the average
Here is the part almost nobody mentions. People left these trials, and the averages were computed on those who remained. In Zepbound Study 1, the proportions who discontinued study drug were 14.3, 16.4 and 15.1 percent on the 5, 10 and 15 mg doses, against 26.4 percent on placebo. In the Wegovy maintenance trial, of 902 patients who began a 20-week run-in, 99, or 11 percent, discontinued before randomization, most commonly for adverse reactions, 48 patients, or 5.3 percent.
That is not an indictment of the medicines; a quarter of the placebo group left too. It is arithmetic honesty about the mean. A number that summarizes the people who stayed cannot promise anything to a person who might not have stayed, and nobody knows in advance which kind they will be. The label reports who achieved thresholds because the spread is real; it does not report who the average would have described had everyone remained, because nobody can.
Time is the misread variable
The second quiet trap in the headline numbers is duration. Those figures belong to 68 and 72 week trials with a year or more at the maintenance dose. A goal set in week eight, and judged in week twelve, is being measured against a number earned over roughly eighteen months. The arc of a typical first year, including the stretch where loss slows for physiological reasons, is its own subject at Month by Month: What GLP-1 Results Actually Look Like.
This is where goals go wrong most often, and where revising one gets misread as failing one. Weight loss on these medicines typically slows as the body adapts, which is expected physiology rather than a verdict, covered at Why Weight Loss Always Slows Down and Hit a Plateau on a GLP-1? What Your Data Can Tell You. A target calibrated to month three of a fast start was never calibrated to month fourteen of anything. Adjusting it as the timeline reveals itself is what a careful person does, not an admission of defeat.
Where a target comes from, and what to watch instead
So where should a number come from? A conversation with a prescriber, who can weigh a starting point, other conditions, and what the medicine is being prescribed for in the first place, since these products carry different approved uses and the populations in their trials differ accordingly. This article will not name a target, a percentage or a BMI, because a defensible one cannot come from an article; it comes from that conversation, revised as real data accumulates.
Meanwhile, the things worth watching are bigger than a single number. A trend line rather than any one reading, explained at How to Read a Weight Trend Line. Measurements that a scale cannot see, at Beyond the Scale: Measurements Worth Tracking on a GLP-1. The markers of progress that have nothing to do with kilograms at all, at Non-Scale Victories: The Progress Your Scale Misses. A goal set as a single morning reading invites a verdict every single day; a goal kept as a direction, backed by a trend, survives bad weeks.
And when the number finally does arrive, whatever it turns out to have been, the story does not end there. What changes after a target is reached is the subject of Maintenance Mode: Tracking After You Reach Goal Weight, and it is the honest punchline of this whole subject: most people who get there report that the target was less important than what they built on the way, the data, the habits, the record of how it actually went.
Frequently asked questions
How much weight will I lose on a GLP-1?
No honest answer exists at the level of the individual, and the labels say so in their own way. The Zepbound trials reported group averages like a 17.7 and 19.6 percent reduction against placebo in sleep apnea studies, and they also published how many people reached 5, 10, 15 and 20 percent, precisely because the average describes nobody. Substantial numbers of participants land above the mean and substantial numbers below it. Your own trajectory is something you observe with your clinician, not a number you inherit from a trial.
What is a good goal weight to aim for on Ozempic or Wegovy?
This article deliberately names no target, no percentage and no BMI, because a goal weight is not a number you pick from an article. It belongs in a conversation with a prescriber, who can weigh your starting point, other conditions, and what the medicine is being prescribed for in the first place. What an article can do is explain how to read the trial figures honestly: as group averages over 68 and 72 weeks, computed on the people who stayed in the trials.
Why does my weight loss not match the percentage from the trials?
Several ordinary reasons. Trial averages are the center of a wide spread, with substantial fractions of participants above and below them. They were earned over 68 and 72 weeks, with a year or more at the maintenance dose, so an early-month comparison is against a number that took eighteen months. And the averages were computed on the people who stayed, while a meaningful share left the trials, sometimes for side effects. None of this says anything is wrong; it says the headline number was never a promise.
Is it normal to change my goal weight partway through?
Yes, and revising a goal is not failure. As weight comes down the body adapts and loss slows for physiological reasons, which is covered in the plateau and metabolic adaptation pages on this site. A target set before starting, with no lived experience of the medicine, is a guess by definition. People commonly adjust goals in either direction with their prescriber as real data accumulates, and the data, a trend line rather than single readings, is what informs that conversation.
What should I track instead of just a goal number?
A reading is an event, a trend line is information. GLP 1 Tracker logs weight day by day and keeps progress photos, so what you have after a month is a run of readings rather than one morning's number. The markers a scale cannot show belong in the record too: how clothes fit, stairs, energy, the things the non-scale victories page on this site describes. When the goal number finally arrives, the maintenance page covers what changes, and most people find the target mattered less than what got built on the way.