GLP-1 Weight Loss Per Week: Averages From the Trials
GLP-1 weight loss per week is usually quoted without any source. This page derives it transparently: the two verified trial totals divided by their trial lengths, labeled as rough averages across the whole trial, with the real pattern, faster in the middle and slower at both ends, explained. It is education only, not medical advice.
The two verified totals this page starts from
Every weekly number on this page is arithmetic on two published trial results, and nothing else. In STEP 1, participants on semaglutide 2.4 mg lost an average of 14.9 percent of their body weight at 68 weeks. In SURMOUNT-1, participants on tirzepatide at the 15 mg dose lost an average of 20.9 percent of their body weight at 72 weeks. These figures come from separate trials and are not a head to head comparison: different medicines, different groups, different durations. They are placed side by side here only because both are needed to answer the per week question honestly.
Notice what the totals already tell you before any division happens. Both were measured after well over a year of treatment, including months of dose escalation at the start. Any weekly figure derived from them is therefore an average across a long, uneven process, not a rate anyone experienced week after week.
The arithmetic, shown openly
Dividing the STEP 1 total by its length gives 14.9 divided by 68, which is about 0.22 percent of starting body weight per week. Dividing the SURMOUNT-1 total by its length gives 20.9 divided by 72, which is about 0.29 percent of starting body weight per week. Those are the rough weekly averages across the whole trial for each study, and they are the only weekly rates this page will state, because they are the only ones the evidence supports.
Percentages keep the numbers honest across different starting weights, but an example helps them land. For a person starting at 100 kilograms, the STEP 1 rough average corresponds to about 0.22 kilograms per week across the trial, and the SURMOUNT-1 rough average to about 0.29 kilograms per week. For a different starting weight, scale the percentage, not the kilograms: the percentage is the trial figure, the kilograms are just that percentage applied to one example body.
| Trial | Total average loss | Trial length | Rough average per week | Rough average per month |
|---|---|---|---|---|
| STEP 1, semaglutide 2.4 mg | 14.9 percent | 68 weeks | About 0.22 percent | About 0.9 percent |
| SURMOUNT-1, tirzepatide 15 mg | 20.9 percent | 72 weeks | About 0.29 percent | About 1.2 percent |
The monthly column is the same division carried one step further, the weekly average multiplied up to a month, and it carries the same warning label: rough average across the whole trial. It is useful for judging a monthly trend. It is meaningless as a weekly expectation.
Why real weeks do not look like the average
Loss on these medicines is usually slower at the start, faster in the middle, and slower again at the end, and the average hides all three phases. The start is slow because both schedules spend months rising from a low starting dose to maintenance, so early weeks are counted in the average at doses well below the one the total was measured on. The middle, at a stable maintenance dose, is where most people do better than the average line. The end slows because a lighter body needs less energy, adaptation pushes back, and plateaus arrive.
Week to week readings add their own noise on top. Fluid shifts, constipation from slower digestion, the day of the week you weigh, and where you are in the dose cycle can each move a single reading more than a full week of average loss. That is why this site recommends weighing no more than weekly under consistent conditions and judging the monthly trend against the rough monthly averages above, rather than auditing every week against 0.22 or 0.29 percent.
A worked example at two starting weights
To see why the percentage form matters, apply the same rough averages to a second example. A person starting at 80 kilograms would, on the STEP 1 rough average, be looking at about 0.18 kilograms per week across the trial, and on the SURMOUNT-1 rough average at about 0.23 kilograms per week. A person starting heavier loses more kilograms at the same percentage, and a person starting lighter loses fewer, while the percentage pace, the only pace the trials actually measured, is identical. This is also why comparing kilograms per week with a friend on the same medicine misleads: unless your starting weights match, you are comparing two different scales dressed as one number.
The same logic protects you in the other direction. A heavier person reporting a large weekly kilogram loss online may simply be expressing the ordinary percentage average on a larger starting weight. The percentage is the transferable figure. Everything else is that figure wearing the clothes of one particular body at one particular starting point.
How to use these averages well
Used properly, the per week figures do two jobs. First, they set scale: they show that the celebrated trial results, spread honestly, are a fraction of a percent of body weight per week, which reframes the first month completely. A small early loss is not behind schedule; the schedule itself is gradual. Second, they give a fair long term check: after escalation, a monthly trend somewhere near the rough monthly average, above it or below it, is within the normal landscape, because the endpoint is an average of people who finished above and below it.
Used badly, the same figures become a weekly pass and fail test that almost nobody would pass consistently, including many trial participants themselves. If your trend sits below the line for a long stretch at a stable dose, that is information for a prescriber review, alongside dose stage, protein, sleep, activity and other medicines, not a verdict to deliver to yourself at the scale. And if a source quotes you a faster guaranteed weekly rate than simple division of the trial totals produces, ask what evidence it is dividing, because there is no larger trial total to divide.
Frequently asked questions
What is the average GLP-1 weight loss per week in the trials?
As rough arithmetic across the whole trial, STEP 1 works out to about 0.22 percent of starting weight per week, from 14.9 percent over 68 weeks, and SURMOUNT-1 to about 0.29 percent per week, from 20.9 percent over 72 weeks. These are separate trials, not a head to head comparison.
Why is my weekly loss slower than the average some weeks?
The average spreads a long, uneven process evenly. Real loss is usually slower during dose escalation at the start, faster in the middle at maintenance, and slower again near the end, with plateaus and fluid noise in between. A slow week is normally part of the pattern, not a departure from it.
How many kilograms per week is the trial average?
It depends on starting weight, because the trial figures are percentages. For a 100 kilogram start, the rough averages correspond to about 0.22 kilograms per week for STEP 1 and about 0.29 kilograms per week for SURMOUNT-1, across the whole trial period.
Is faster weekly loss better on a GLP-1 medicine?
Not necessarily, and this page sets no target rate. The trial averages were achieved gradually over more than a year, and very rapid loss from any cause raises concerns such as muscle loss. Your prescriber judges your pace in the context of your dose, health and side effects.
Can I use the weekly average to predict my result?
No. The averages describe groups over full trial periods and are useful for scale and for judging a monthly trend. Individual results vary widely around them, and no weekly figure can predict what a particular person will lose in a particular week or in total.