Percent Weight Loss vs Pounds Lost GLP-1 Studies: What the Numbers Actually Mean
Every major GLP-1 drug, every label-reported dose, both populations — FDA-label percent converted to trial-average pounds and your-weight pounds at 220 lb. Plus: why headline numbers and FDA-label numbers often don't match.
Published: · Last reviewed:
The 30-Second Answer
Here's the math at the four most-searched starting weights, before we explain anything:
| Percent loss | At 180 lb | At 220 lb | At 260 lb | At 300 lb |
|---|---|---|---|---|
| 5% | 9 lb | 11 lb | 13 lb | 15 lb |
| 10% | 18 lb | 22 lb | 26 lb | 30 lb |
| 15% | 27 lb | 33 lb | 39 lb | 45 lb |
| 20% | 36 lb | 44 lb | 52 lb | 60 lb |
| 22.5% | 41 lb | 50 lb | 59 lb | 68 lb |
| 25% | 45 lb | 55 lb | 65 lb | 75 lb |
The formula:
Pounds lost = starting weight × percent ÷ 100
Example: 15% of 220 lb = 220 × 15 ÷ 100 = 33 lb
The rest of this page is about which percentage you should be plugging in — because the wrong one will give you a number that looks real but isn't.
How Do I Convert Percent Weight Loss to Pounds?
Multiply your starting weight (in pounds) by the percent, then divide by 100. That's it. 15% of 220 lb = 220 × 15 ÷ 100 = 33 lb. The reverse works too: if you've lost 33 lb from 220 lb, that's 33 ÷ 220 × 100 = 15%.
Percent → Pounds
weight × percent ÷ 100
Pounds → Percent
pounds ÷ weight × 100
Why “15%” Doesn't Mean “15 Pounds”
A percentage is a fraction of something, not a fixed amount. Fifteen percent of a small number is small. Fifteen percent of a big number is big. That's why two people on the same drug can both lose “15%” and one drops 27 pounds while the other drops 45.
The Most Common Mistake
People hear “Wegovy users lost 15%,” see Reddit posts that say “I lost 80 pounds,” and assume something is off. Usually nothing is off. Eighty pounds at 350 lb is 22.8%. Fifteen percent at 320 lb is 48 pounds. The numbers aren't fighting — they're describing different people with different starting weights at different points in their treatment.
Why Do GLP-1 Studies Report Percent Instead of Pounds?
Because FDA guidance evaluates weight-loss drugs in percent. The FDA's January 2025 draft guidance — nonbinding and labeled “Not for implementation” — lists mean percentage change in body weight as the adult primary efficacy endpoint and says that, in general, a drug is considered effective if the treatment-vs-control difference is statistically significant and at least 5% after one year at the maintenance dose.
Reason 1: Percentages compare fairly across different bodies
A 30-pound loss is dramatic in a 180-pound person and modest in a 350-pound person. If a trial reports results in pounds, the heaviest people drag the average up. Reporting in percent fixes that. A 15% loss means the same share of starting weight at every body size.
Reason 2: Trials and labels track outcomes by percent
Obesity studies and FDA labels evaluate weight outcomes by percent body weight change because that's the unit that makes results comparable across people of very different starting weights. The specific health benefit at any given percent depends on the person and the condition, which is one reason individual care plans matter.
But pounds still matter to you — for a different reason
Pounds are what the scale shows. Pounds are what your jeans tell you. Pounds are what your brain uses to feel progress. Use percent to compare between drugs and studies. Use pounds to picture your real life.
Percent Weight Loss vs Pounds Lost GLP-1 Studies: What the Major Trials Show
Across FDA-approved labels, average weight loss ranged from about 11.5 lb (Foundayo 5.5 mg in the type 2 diabetes population) to about 48.7 lb (Zepbound 15 mg in the no-diabetes population) at trial endpoints of 56 to 72 weeks.
The GLP-1 Percent-to-Pounds Study Matrix
Every percentage and baseline weight is pulled from the current FDA/DailyMed prescribing labels. Trial baseline weights converted from kg to lb (1 kg = 2.20462 lb). When press releases and FDA labels report different numbers from the same trial, the label number takes precedence in our matrix. Last verified May 16, 2026.
| Drug / dose | Population | Weeks | Baseline | Label % | Trial avg lb | At 220 lb | Placebo | Source |
|---|---|---|---|---|---|---|---|---|
| Wegovy 2.4 mg (injection) | No type 2 diabetes | 68 | 232.4 lb | −14.9% | 34.6 lb | 32.8 lb | −2.4% / 5.6 lb | Wegovy label (STEP 1) |
| Wegovy 2.4 mg (injection) | Type 2 diabetes | 68 | 220.2 lb | −9.6% | 21.1 lb | 21.1 lb | −3.4% / 7.5 lb | Wegovy label (STEP 2) |
| Wegovy 25 mg (oral tablet) | No type 2 diabetes | 64 | 234.6 lb | −13.6% | 31.9 lb | 29.9 lb | −2.4% / 5.5 lb | Wegovy label (OASIS 4) |
| Wegovy HD 7.2 mg | No type 2 diabetes | 72 | 247.8 lb | −18.8% (label) / −20.7% (efficacy) | 46.6 / 51.3 lb | 41.4 / 45.5 lb | −3.9% / 9.7 lb | Wegovy HD label (STEP UP) |
| Wegovy HD 7.2 mg | Type 2 diabetes | 72 | 243.6 lb | −13.2% | 32.2 lb | 29.0 lb | −3.8% / 9.4 lb | Wegovy HD label (STEP UP T2D) |
| Drug / dose | Population | Weeks | Baseline | Label % | Trial avg lb | At 220 lb | Placebo | Source |
|---|---|---|---|---|---|---|---|---|
| Zepbound 5 mg | No type 2 diabetes | 72 | 226.9 lb | −15.0% | 34.0 lb | 33.0 lb | −3.1% / 7.2 lb | Zepbound label (SURMOUNT-1) |
| Zepbound 10 mg | No type 2 diabetes | 72 | 233.2 lb | −19.5% | 45.5 lb | 42.9 lb | −3.1% / 7.2 lb | Zepbound label (SURMOUNT-1) |
| Zepbound 15 mg | No type 2 diabetes | 72 | 232.8 lb | −20.9% | 48.7 lb | 46.0 lb | −3.1% / 7.2 lb | Zepbound label (SURMOUNT-1) |
| Zepbound 10 mg | Type 2 diabetes | 72 | 222.4 lb | −12.8% | 28.5 lb | 28.2 lb | −3.2% / 7.2 lb | Zepbound label (SURMOUNT-2) |
| Zepbound 15 mg | Type 2 diabetes | 72 | 219.6 lb | −14.7% | 32.3 lb | 32.3 lb | −3.2% / 7.2 lb | Zepbound label (SURMOUNT-2) |
| Drug / dose | Population | Weeks | Baseline | Label % | Trial avg lb | At 220 lb | Placebo | Source |
|---|---|---|---|---|---|---|---|---|
| Foundayo 5.5 mg (oral) | No type 2 diabetes | 72 | 227.5 lb | −7.4% | 16.8 lb | 16.3 lb | −2.1% / 4.8 lb | Foundayo label (ATTAIN-1) |
| Foundayo 9 mg (oral) | No type 2 diabetes | 72 | 225.3 lb | −8.3% | 18.7 lb | 18.3 lb | −2.1% / 4.8 lb | Foundayo label (ATTAIN-1) |
| Foundayo 17.2 mg (oral) | No type 2 diabetes | 72 | 227.3 lb | −11.1% | 25.2 lb | 24.4 lb | −2.1% / 4.8 lb | Foundayo label (ATTAIN-1) |
| Foundayo 5.5 mg (oral) | Type 2 diabetes | 72 | 225.5 lb | −5.1% | 11.5 lb | 11.2 lb | −2.5% / 5.6 lb | Foundayo label (ATTAIN-2) |
| Foundayo 9 mg (oral) | Type 2 diabetes | 72 | 226.4 lb | −7.0% | 15.8 lb | 15.4 lb | −2.5% / 5.6 lb | Foundayo label (ATTAIN-2) |
| Foundayo 17.2 mg (oral) | Type 2 diabetes | 72 | 220.0 lb | −9.6% | 21.1 lb | 21.1 lb | −2.5% / 5.6 lb | Foundayo label (ATTAIN-2) |
Note: Foundayo's approved escalation also includes 14.5 mg, but FDA-label efficacy tables report results for 5.5 mg, 9 mg, and 17.2 mg only. We do not invent a 14.5 mg estimate.
| Drug / dose | Population | Weeks | Baseline | Label % | Trial avg lb | At 220 lb | Placebo | Source |
|---|---|---|---|---|---|---|---|---|
| Saxenda 3.0 mg (injection) | No type 2 diabetes | 56 | 234.1 lb | −7.4% | 17.3 lb | 16.3 lb | −3.0% / 7.0 lb | Saxenda label (SCALE Obesity) |
| Saxenda 3.0 mg (injection) | Type 2 diabetes | 56 | 233.0 lb | −5.4% | 12.6 lb | 11.9 lb | −1.7% / 4.0 lb | Saxenda label (SCALE Diabetes) |
Two Things to Notice Before You Keep Reading
First: every drug shows two rows. No type 2 diabetes, and type 2 diabetes. The same drug at the same dose consistently produces less average percent loss in trials of people with type 2 diabetes. Wegovy 2.4 mg: 14.9% becomes 9.6%. Zepbound 15 mg: 20.9% becomes 14.7%. Foundayo 17.2 mg: 11.1% becomes 9.6%. If you have diabetes, the diabetes row is the one that fits your situation, not the headline.
Second: notice the Wegovy HD line has two percentages. That's not a typo. The FDA label and the marketing press release report different numbers for the same trial. We explain why in the next section.
Zepbound 15 mg shows the highest trial average: 48.7 lb at 72 weeks.
Ro and LillyDirect both carry the full Zepbound line at the same cash price. If you have insurance, prior authorization support is available.
Affiliate disclosure: we may earn a commission. That does not change the label data on this page.
Headline Numbers vs FDA-Label Numbers: Where They Diverge
The same trial can produce two correct percentages depending on which statistical analysis you read. Press releases tend to publish the higher one. FDA labels tend to publish the more conservative one. Both are real. Most consumer pages quote the press-release number without explaining the difference.
| Headline / claim | Where it comes from | Claimed | FDA label | Why they differ | Which to use |
|---|---|---|---|---|---|
| "Zepbound delivers 22.5% weight loss" | Lilly SURMOUNT-1 press release, 15 mg | −22.5% | −20.9% | Press release uses efficacy estimand (adherent participants); label uses treatment-regimen (everyone randomized) | Label number for personal expectations |
| "Wegovy HD: 20.7% mean weight loss" | Novo Nordisk STEP UP press release | −20.7% | −18.8% | Same estimand split as above | Label number for personal expectations |
| "Wegovy reduces body weight by 15%" | STEP 1 / Wegovy 2.4 mg, no-diabetes | −14.9% | −14.9% | Same number, same study — but applies only to the no-diabetes population | This row only if you don't have type 2 diabetes |
| "GLP-1s produce ~20% weight loss" | General headline | ~20% | −5.1% to −20.9% | Headline blends multiple drugs, doses, populations, and estimands | The specific matrix row that matches your situation |
What's an Estimand, in Plain English?
An estimand is a precise definition of what a clinical trial is actually measuring. There are two main flavors in obesity trials:
Treatment-Regimen Estimand
What happened to everyone randomized, including people who quit or used rescue treatments. More conservative. Closer to real life. FDA labels usually lead with this.
Efficacy Estimand
What happened to people who stayed on the drug and followed the protocol. Higher number. Press releases prefer this one. The 1.5- to 2-point gap between them is the difference between the FDA label result and the marketing number.
Same trial. Same patients. Same data. Different question. Different answer.
Why Do Some GLP-1 Studies Show Different Numbers for the Same Medication?
Because percentages depend on five hidden variables: dose, study population, endpoint duration, raw vs placebo-adjusted, and which estimand the analysis used. Two reputable sources can publish two different numbers for the same drug and both be correct — they're just measuring different things.
Dose changes the number
Higher dose, more loss. Zepbound 5 mg gives about 15%. Zepbound 10 mg gives about 19.5%. Zepbound 15 mg gives about 20.9%. Same drug, same trial, three different headlines. If someone tells you "Zepbound gives 22.5%," they're talking about the highest dose under best-case analysis, not the average dose your prescriber may actually start you on.
Population changes the number
Type 2 diabetes trial populations show smaller average percent loss than no-diabetes obesity trial populations across every GLP-1 drug. Don't compare a no-diabetes Wegovy result to your diabetes situation. Use the diabetes row.
Endpoint duration changes the number
Saxenda's trials reported 56-week results. Wegovy and Zepbound reported 68- and 72-week results. Foundayo reports 72 weeks. A longer trial captures more weight loss because most people are still losing at 6 months. If you compare a 56-week number to a 72-week number, the 72-week drug looks better than it really is in practice.
Raw loss vs placebo-adjusted loss
The drug group in the Wegovy 2.4 mg no-diabetes trial lost 14.9%. The placebo group lost 2.4%. The placebo-adjusted difference is 12.4 percentage points. Both numbers are real. They answer different questions.
Estimand changes the number — and this is the big one
The label leads with one estimand; press releases lead with another. The numbers can differ by 1 to 3 percentage points for the same drug at the same dose in the same trial. See the divergence table above.
Should I Use Raw Pounds Lost or Placebo-Adjusted Pounds Lost?
Use the raw number when you want to know what the scale might show. Use the placebo-adjusted number when you want to know how much more the medication group lost than the placebo/lifestyle group. They're not interchangeable, and most online sources blur them.
Raw Number Answers:
“What did people taking the drug actually lose?”
In the Wegovy 2.4 mg no-diabetes label trial, people on the drug lost an average of 14.9% of body weight — about 34.6 pounds on the scale at the 232-pound trial baseline. Use this to picture your scale at month 12.
Placebo-Adjusted Answers:
“How much more than the placebo group?”
In the same trial, placebo lost 2.4% (5.6 lb). The placebo-adjusted difference is 12.4 percentage points. Use this to evaluate whether the drug arm beat the placebo arm by a meaningful amount in trial conditions.
Wegovy HD has the largest placebo-adjusted gap in our matrix: ~14.9 pp in the no-diabetes data. Saxenda has the smallest: ~4.4 pp in its no-diabetes data.
A bigger placebo-adjusted gap means the drug arm outperformed the placebo arm by more under that trial's conditions. A smaller gap means the two arms were closer.
For a deeper look at this distinction, see our Placebo-Adjusted Weight Loss Explained guide.
Should You Use the No-Diabetes Row or the Type 2 Diabetes Row?
Use the type 2 diabetes row if you have type 2 diabetes. Use the no-diabetes row if you don't. The gap between them is bigger than most readers realize — usually 2 to 6 percentage points in the same drug at the same dose.
What the Gap Looks Like, Drug by Drug
| Drug / dose | No diabetes | With diabetes | Gap |
|---|---|---|---|
| Wegovy 2.4 mg | −14.9% | −9.6% | 5.3 pp |
| Wegovy HD 7.2 mg | −18.8% (label) | −13.2% | 5.6 pp |
| Zepbound 15 mg | −20.9% | −14.7% | 6.2 pp |
| Foundayo 17.2 mg | −11.1% | −9.6% | 1.5 pp |
| Saxenda 3.0 mg | −7.4% | −5.4% | 2.0 pp |
Why the gap exists
Type 2 diabetes affects how the body stores and burns energy. People with type 2 diabetes often have insulin resistance, which makes weight loss biologically harder than for people without diabetes, even at the same calorie deficit. Some diabetes medications — especially insulin and certain sulfonylureas — also promote weight gain, which can offset some of the GLP-1's effect.
One important note
This isn't a reason to write off GLP-1 medications if you have type 2 diabetes. For people with diabetes, these drugs also improve A1C and reduce cardiovascular risk in many studies — health benefits the scale doesn't show. The right comparison isn't “9.6% vs 14.9%.” It's “9.6% plus diabetes improvements vs no treatment.”
Which GLP-1 Study Row Should I Use for My Own Starting Weight?
Use the row that matches your medication, dose, diabetes status, and endpoint duration. If your situation doesn't match a row exactly, use the matrix as context — not a personal forecast.
Decision filter
- →Match your medication and dose. A 5 mg Zepbound user shouldn't use the 15 mg row. Use the row for the dose you're actually taking or planning to titrate up to.
- →Match your diabetes status. No type 2 diabetes → no-diabetes row. Type 2 diabetes → diabetes row.
- →Don't extrapolate from a higher dose. If side effects keep you at 5 mg Zepbound, expect results closer to the 5 mg row, not the 15 mg headline.
- →Don't judge your first month against a 72-week endpoint. The percentages on this page are at trial end, not at week 4.
If none of the rows match — different drug, off-label use, compounded product without trial data — use the closest analog in the matrix as context only and have an honest conversation with your prescriber about realistic expectations.
Why Does Reddit Make It Look Like Everyone Loses Way More Than the Studies?
Three reasons: survivor bias, late-stage timing, and starting-weight skew. The average Reddit post is not the average trial participant. People who lost 80 pounds post about it. People who lost 12 pounds quietly continue their lives. Many of the dramatic posts are from people 12 to 18 months in, at maintenance dose, who started at higher weights.
| Reddit phrase | What the study math says |
|---|---|
| "I lost 80 lb on a GLP-1" | At 350 lb start, that's 22.9% — within the top-responder range for SURMOUNT-1 15 mg. At 250 lb start, it's 32% — well above any trial average and likely 18+ months in. |
| "20 lb a month" | Possible during the first 1–3 months at high starting weights, almost never sustained for 72 weeks. Trial-average loss spreads ~33–50 lb across 16+ months. |
| "I'm only losing 2 lb a month" | At 220 lb, that's about 11% over 12 months — close to or slightly below average for Wegovy 2.4 mg and within range of label data. |
| "Why is everyone losing more than the studies?" | They aren't. The posters are a self-selected sample, not the trial average. |
Reddit is useful for finding real questions and real side-effect experiences. It's not a reliable benchmark for what you should expect. For what real-world data actually shows, see our GLP-1 Real-World Evidence vs Clinical Trial Results guide.
Trial Pounds vs Real-World Pounds — What's the Actual Gap?
Real-world GLP-1 results depend heavily on whether the study counts only people who stay on therapy. Persistent-user cohorts can land close to trial figures. All-starter cohorts that include people who discontinue land lower.
Persistent-User Real-World (SHAPE Study)
- 14.1% on semaglutide 2.4 mg at 1 year
- 16.5% on tirzepatide at 1 year
- No-diabetes adults, no treatment gap >30 days. Close to corresponding trial figures for adherent participants.
All-Starter Real-World
When you include people who stop early — roughly half of GLP-1 starters stop within a year due to side effects, cost, supply gaps, or insurance issues — average results drop sharply. People who stop at month 4 do not generally retain a month-12 trial result.
Key takeaways
- If you stay on the drug at full dose for a year, expect something close to the trial figure for your row.
- If you're likely to face cost, supply, or tolerance issues, build in some buffer.
- The single biggest predictor of outcome isn't genetics or starting weight. It's whether people stay on the medication consistently.
The Non-Responder Reality
Some people lose very little even at full dose. A 2026 Nature genetic study of 27,885 self-reported GLP-1 users found that GLP1R and GIPR gene variants are linked to differences in weight-loss response — but the practical signal was modest: about 0.76 kg of additional weight loss per copy of the lead GLP1R variant. Response varies for biological reasons too, and averages hide that spread.
For more on why people stop GLP-1 treatment and how that affects outcomes, see our GLP-1 Trial Dropout Rates Explained guide.
Is Tirzepatide Actually a GLP-1?
Tirzepatide (the active ingredient in Zepbound and Mounjaro) is commonly grouped with GLP-1 medications, but technically it's a dual GIP/GLP-1 receptor agonist — it activates two hormone receptors instead of one (per the Zepbound prescribing information). That second receptor (GIP, or glucose-dependent insulinotropic polypeptide) appears to be why tirzepatide tends to outperform semaglutide on average in head-to-head trials.
SURMOUNT-5: The Direct Head-to-Head (NEJM, May 2025)
751 adults randomized to maximum-tolerated tirzepatide (10 or 15 mg) vs maximum-tolerated semaglutide (1.7 or 2.4 mg) for 72 weeks:
Tirzepatide
−20.2%
~50 lb at trial mean
Semaglutide
−13.7%
~33 lb at trial mean
6.5-pp gap, ~17 more pounds at the trial mean baseline. When compared at maximum tolerated dose under identical conditions, tirzepatide wins on average. Individual response still varies — some people respond better to semaglutide.
Throughout this page, “GLP-1 studies” loosely includes both pure GLP-1 receptor agonists (semaglutide, liraglutide, orforglipron) and dual GIP/GLP-1 agonists (tirzepatide). That's how searchers ask the question. We keep the pharmacology accurate when it matters.
How Long Does It Take to Lose That Weight?
The trial percentages everyone quotes come from endpoints of 56 to 72 weeks — about 13 to 17 months. None of those numbers describe the first month, or even the first quarter.
| Time on drug | Typical mean loss (no diabetes) | Notes |
|---|---|---|
| Month 1 | 1–3% | Mostly during dose escalation, often includes water weight |
| Month 3 | 5–7% | First "clinically meaningful" milestone |
| Month 6 | 9–12% | Most people are at or near maintenance dose |
| Month 9 | 12–17% | Loss curve starts to flatten for some |
| Month 12 | 14–20% | Approaching individual plateau |
| Month 17 | 15–22.5% | Trial endpoint range |
Early responders
Lose noticeable weight in the first 12 weeks and tend to keep going. In SURMOUNT-1, people who hit ≥5% loss by week 12 were highly likely to hit ≥15% by week 72.
Late responders
Lose slowly through month 6 and accelerate after. A SURMOUNT-1 post-hoc analysis found 90% of “late responders” still reached at least 5% loss by week 72.
If you're two months in and not seeing Reddit-grade results
You're not failing — you're being normal. Don't write off a 72-week endpoint at week 8. If weight loss is slow or side effects are pushing you to stop, that's a conversation with your prescriber, not a verdict.
What Can Make Your Actual Pounds Lost Different from the Study Average?
Five biggest levers, ranked roughly by impact: adherence, dose reached, starting weight, diabetes status, and individual biological response. The study average assumes you're at maintenance dose for the full duration and not dropping out — which most people in the real world don't pull off.
Adherence
The single biggest predictor. Trial participants who took ≥80% of doses got the trial-average result. Persistent users in real-world SHAPE data landed close to trial figures. People who stopped within six months in other real-world cohorts averaged only a few percent of body weight loss.
Dose reached
A 5 mg Zepbound user shouldn't expect a 15 mg Zepbound result. The label percentages are for the labeled maintenance dose. If side effects keep you at a lower dose, expect a result closer to the lower-dose row.
Starting weight
Higher starting weight produces more pounds at the same percentage. A 320 lb person losing 15% loses 48 pounds. A 180 lb person losing 15% loses 27 pounds. The percentage may be similar; the pound experience is very different.
Diabetes status
Already covered above. Use the diabetes row if it applies to you.
Individual biological response
People vary widely. Top responders in SURMOUNT-1 lost more than 25% of body weight; others on the same drug at the same dose lost much less. You won't know which group you're in until you try, which is why prescribers usually evaluate response at 3 to 6 months and adjust.
Smaller Modifiers Worth Noting
- Sex. A 2026 Johns Hopkins meta-analysis (excluding tirzepatide by design) found women averaged about 10.88% weight loss vs about 6.78% for men on GLP-1 receptor agonist trials. No clear systematic differences were found across age, race, ethnicity, initial BMI, or baseline HbA1c.
- Lifestyle support. Trials included structured counseling; real-world prescribing usually doesn't. People who add nutrition or exercise support tend to do better.
- Stopping the drug. In the STEP 1 extension, people who stopped semaglutide 2.4 mg after 68 weeks regained about two-thirds of their prior weight loss during the following year off treatment. In a SURMOUNT-4 post-hoc analysis, 82% of participants switched to placebo regained more than 25% of the weight they had lost within one year. GLP-1s appear to be a long-term treatment, not a short-term reset.
What About Body Composition? Is GLP-1 Weight Loss All Fat?
The scale measures total body weight. Trial weight loss includes fat, lean tissue, water, and gut contents — not just fat. In the SURMOUNT-1 DXA substudy, about 75% of weight lost on tirzepatide was fat mass and about 25% was lean mass on average. That ratio is similar to other weight-loss methods, including dieting.
| Total scale loss | Approx. fat mass (~75%) | Approx. lean mass (~25%) |
|---|---|---|
| 20 lb | ~15 lb | ~5 lb |
| 40 lb | ~30 lb | ~10 lb |
| 60 lb | ~45 lb | ~15 lb |
Based on the SURMOUNT-1 DXA substudy group average — not an individual body composition test. Your actual split could differ based on protein intake, resistance training, and other factors.
- • A 40-pound scale loss is not a 40-pound fat loss.
- • Resistance training and a protein target set with your clinician can help protect lean mass during weight loss.
- • The percentages in our matrix are total body weight changes from FDA labels — they are not fat-loss percentages.
- • If body composition matters to you, ask your prescriber about a DEXA scan before starting and one near month 6 or month 12.
How Should I Use These Study Numbers Before Talking to a Clinician?
Pick the row that matches your situation — your drug, your dose, your diabetes status — then translate the percentage using your own starting weight. Bring that pound range to your clinician as a starting point for the conversation, not as a target you've committed to.
Script to use with your prescriber
“I read the label data for [drug] at the [dose] in the [no-diabetes / type 2 diabetes] population. The average percent loss at 72 weeks was [X]%. At my starting weight of [your weight] pounds, that translates to about [Y] pounds. Is that a reasonable range for me to expect, given my situation? What would change it up or down?”
Questions Worth Bringing to a Clinician
- →Which dose are you starting me at, and what's the timeline to maintenance dose?
- →Should I use the no-diabetes or type 2 diabetes row to set my expectations?
- →What's your plan if I'm not seeing 3–5% loss by month 3?
- →What side effects should make us change dose or stop?
- →What labs are we monitoring (especially if I have diabetes)?
- →What happens to my weight if I stop the drug at some point?
- →How will we define success beyond pounds?
Ready to have this conversation with a prescriber?
Sesame connects you with licensed clinicians who can evaluate your situation, prescribe if appropriate, and help you set realistic expectations from the data above.
Affiliate disclosure: we may earn a commission. That does not change the label data on this page.
How We Built and Verified This Page
We pulled every percentage and baseline weight in our matrix directly from the current FDA/DailyMed prescribing labels — not from press releases, not from secondary summaries. We then ran every pound conversion using the same formula and rounded to one decimal place.
Source Hierarchy
- FDA/DailyMed prescribing labels for Wegovy, Zepbound, Foundayo, and Saxenda — accessed May 2026.
- Peer-reviewed pivotal trial publications (NEJM, Lancet Diabetes & Endocrinology) when label tables didn't break down a specific subgroup.
- FDA approval announcements for newly approved drugs (Foundayo April 1, 2026; Wegovy HD March 19, 2026).
- Manufacturer press releases (Lilly, Novo Nordisk) used for context only. When press releases and FDA labels report different numbers from the same trial, label numbers take precedence.
- Peer-reviewed real-world cohort studies (SHAPE; discontinuation analyses) for the real-world section — clearly separated from trial data.
Calculation Method
- Baseline lb = baseline kg × 2.20462
- Trial-avg lb = baseline lb × label percent ÷ 100
- "At 220 lb" = 220 × label percent ÷ 100
- Placebo lb = placebo baseline × placebo percent ÷ 100
What We Actually Verified
- ✅ Wegovy 2.4 mg (STEP 1 / STEP 2) — current Wegovy label
- ✅ Wegovy 25 mg oral (OASIS 4) — current Wegovy label
- ✅ Wegovy HD 7.2 mg (STEP UP) — both −18.8% label and −20.7% press-release numbers, with explanation
- ✅ Zepbound 5/10/15 mg (SURMOUNT-1) and 10/15 mg (SURMOUNT-2) — current Zepbound label
- ✅ Foundayo 5.5/9/17.2 mg (ATTAIN-1 no-diabetes and ATTAIN-2 type 2 diabetes) — current Foundayo label
- ✅ Saxenda 3.0 mg (SCALE) — current Saxenda label
- ✅ SURMOUNT-5 tirzepatide vs semaglutide — NEJM May 2025
- ✅ SHAPE real-world cohort (14.1% semaglutide, 16.5% tirzepatide at 1 year) — peer-reviewed 2025
- ✅ STEP 1 extension regain figure — Wilding et al., extension data
- ✅ SURMOUNT-4 regain statistic — post-hoc analysis published 2025
- ✅ Johns Hopkins meta-analysis sex finding — March 2026, excluding tirzepatide by design
- ✅ SURMOUNT-1 DXA substudy 75% fat / 25% lean split — published 2025
- ⚠️ Nature 2026 GLP1R/GIPR genetic association study — 27,885 self-reported users; the modest effect (~0.76 kg per variant copy) should not be presented as a personal prediction tool
Refresh Cadence
Monthly for the first six months, or immediately if triggered by: new FDA approval, new pivotal trial publication, label update for any drug in the matrix, or a major change in real-world cohort data.
What We Excluded
- • Anecdotal Reddit outcomes as a data source for clinical claims
- • Compounded semaglutide and tirzepatide product claims — not FDA-approved products
- • Off-label uses (alcohol use disorder, PCOS, etc.) — different studies, different endpoints
- • Pediatric weight-loss trial data — separate population requiring its own page
Frequently Asked Questions
How many pounds is 15% body weight loss?
At 180 lb, 15% is 27 lb. At 220 lb, it's 33 lb. At 260 lb, it's 39 lb. At 300 lb, it's 45 lb. The exact number is your starting weight in pounds times 0.15.
How many pounds is 22.5% body weight loss?
At 220 lb, 22.5% is about 50 lb. At 260 lb, about 59 lb. At 300 lb, about 68 lb. 22.5% is the efficacy-estimand figure for Zepbound 15 mg in SURMOUNT-1; the FDA-label treatment-regimen figure is 20.9%, which is the more conservative number.
Why do GLP-1 studies use percentages instead of pounds?
The FDA's January 2025 draft guidance (nonbinding) lists mean percent change in body weight as the adult primary efficacy endpoint for weight-loss drugs, and recommends a 5-percentage-point treatment-vs-control difference as the effectiveness threshold. Percent also lets researchers compare results fairly across people with very different starting weights.
What's the average pounds lost on Wegovy 2.4 mg?
About 34.6 pounds at the no-diabetes trial baseline of 232.4 lb (FDA label: −14.9% over 68 weeks). About 21.1 pounds at the type 2 diabetes trial baseline (−9.6%).
What's the average pounds lost on Zepbound 15 mg?
About 48.7 pounds at the no-diabetes trial baseline (FDA label: −20.9% over 72 weeks). About 32.3 pounds at the type 2 diabetes trial baseline (−14.7%).
What's the average pounds lost on Foundayo at the highest dose?
About 25.2 pounds at the no-diabetes trial baseline of 227.3 lb (FDA label: −11.1% over 72 weeks for the 17.2 mg dose, treatment-policy estimand). About 21.1 pounds at the type 2 diabetes trial baseline (−9.6%).
What's the average pounds lost on Saxenda?
About 17.3 pounds at the no-diabetes trial baseline of 234.1 lb (−7.4% over 56 weeks). About 12.6 pounds at the type 2 diabetes trial baseline (−5.4%). Saxenda has lower label-table weight loss than the highest-dose rows for Foundayo, Wegovy, and Zepbound.
Is 14.9% the same as everyone losing 14.9%?
No. It's a population average. In the Wegovy 2.4 mg no-diabetes label study, 83.5% of participants on the drug lost at least 5%, 66.1% lost at least 10%, and 47.9% lost at least 15%. Some people lost more, some lost less, and some stopped treatment.
Do tirzepatide and semaglutide give the same weight loss?
No, on average. In SURMOUNT-5 (NEJM, May 2025), maximum-tolerated tirzepatide produced 20.2% mean loss vs 13.7% for maximum-tolerated semaglutide over 72 weeks. That's about 17 more pounds at the trial mean baseline. Individual response varies.
Why is the FDA-label percentage lower than the press-release percentage for the same drug?
Most commonly because the label reports the treatment-regimen estimand (everyone randomized, including dropouts) while press releases often report the efficacy estimand (people who stayed on the drug and followed protocol). Both are real numbers from the same trial. The label number is more conservative and closer to real-world conditions.
Should I subtract placebo loss when estimating my own weight loss?
Not for setting personal scale expectations. The raw drug-group result already reflects what happened to people on the drug — including the lifestyle changes baked into the trial. Use placebo-adjusted numbers when evaluating whether the drug arm beat the placebo arm by a meaningful amount, not when predicting your own scale.
How long until I see the trial-average result?
The quoted percentages come from 56- to 72-week endpoints, not early treatment. Month 3 and month 6 results depend on dose escalation, adherence, and the specific drug. Use trial endpoint numbers as context for the long arc — not as a fixed monthly forecast.
What if I have type 2 diabetes — should I expect less weight loss?
On average, yes. Trial data consistently shows smaller percent loss in type 2 diabetes populations across every GLP-1 drug, typically 2 to 6 percentage points lower than the no-diabetes population on the same drug and dose. That said, GLP-1s also produce A1C and cardiovascular benefits in diabetes that the weight number doesn't capture.
Why does Reddit make GLP-1 weight loss look more dramatic than studies?
Success stories cluster online — people who lose 80 pounds post, people who lose 12 don't. Most dramatic posts come from people 12+ months in at maintenance dose, and Reddit posters tend to skew toward higher starting weights where the same percent looks like more pounds. The trial averages are more reliable than the Reddit feed for setting expectations.
The Bottom Line
Compare with percent. Personalize with pounds. Use the FDA-label row that matches your drug, dose, and diabetes status. The headline percentage isn't lying to you — it's just often a different version of the same trial than the one your prescriber's label is based on. Once you know which number you're holding, the rest is multiplication.
Whatever you do next, do it knowing what the number actually means.
This page is educational and does not diagnose, prescribe, or predict your personal medical outcome. GLP-1 and related incretin medications require clinician evaluation and ongoing monitoring. The RX Index is an independent affiliate-supported comparison resource — we may earn commissions when readers connect with telehealth providers we link to, and our editorial choices are not influenced by those relationships. All trial figures pulled from current FDA prescribing labels. Last verified May 16, 2026.