Losing Weight Without a GLP-1: What Actually Works, By the Numbers

You are here because a GLP-1 is not on the table, or not yet. Cost, supply, side effects, a prescriber who said no, or a preference not to inject something weekly. Whatever the reason, the useful thing is a straight comparison of what the alternatives buy you. So here it is, with the numbers the trials produced rather than the numbers the marketing quotes. The short version: the methods differ far less from each other than they differ from a GLP-1, and the thing that separates good outcomes from bad ones is almost never which method you picked.

The honest comparison, first, before anything else

Semaglutide at 2.4 mg produced a mean 14.9% body weight reduction over 68 weeks in its pivotal trial. Behavioural programs with good trial support land in the 3-6% range over six to twelve months. That is not a small gap that a clever protocol closes. It is roughly a threefold difference in outcome, and pretending otherwise is how supplement pages get written.

The gap narrows in one specific place. One year after stopping semaglutide, participants in the STEP 1 extension had regained two thirds of what they lost, finishing 5.6% below their starting weight instead of 17.3% below it. The drug generates the deficit while you take it. What you build alongside it decides where you land afterwards. That is the actual relationship between this section of the site and the medication section, and it is a partnership rather than a competition.

One more piece of context before the numbers. Control groups in lifestyle trials lose weight too, simply from being enrolled, weighed and watched. A program that reports 5% loss in its treatment arm may only own two or three points of that. Read between-group differences where they are available, and treat single-arm headline figures as the ceiling rather than the expectation.

Everything on this page works the same way

Fasting windows, points, colour-coded foods, low carb, low fat, two hard days a week. Each one is a different user interface on the same underlying operation, which is eating less energy than you spend. The trials that matter are the ones that hold calories constant between arms, and those trials keep finding that the differences vanish.

The clearest example is the NEJM trial that gave both groups the same calorie prescription and only varied whether one of them also had to eat inside an eight-hour window. The window added nothing. A network meta-analysis of 99 randomised trials covering 6,582 adults reached the same conclusion across every fasting protocol it examined, with a single exception: alternate-day fasting beat continuous restriction by 1.29 kg, which is a real difference and a modest one.

This matters practically, well beyond the theory. If you believe 16:8 has a metabolic property, you will keep the window and ignore what goes through it, and then you will wonder why the scale stopped. If you understand the window as a way of skipping one eating occasion, you know exactly which lever to pull when progress stalls.

Every method in this cluster, with its published weekly rate

Rates below are percent of body weight per week, converted from trial data at an 85-90 kg starting weight and adjusted against control or placebo where the trial allowed it. Trial conditions run higher than real life, so read the low column as the more likely one. The GLP-1 row sits at the bottom as a reference bar, not as an item to combine with the rest.

Read down the middle column and the point of this whole section becomes obvious in about ten seconds. Almost everything in the behavioural and fasting tiers sits between 0.1% and 0.6% per week. The supplements sit near zero. The differences between methods are smaller than the difference between doing a method and not doing it.

Three tiers, and what separates them

Behavioural programs occupy the top tier because they have the strongest trial support and the widest applicability. The recent digital WeightWatchers RCT in 376 adults produced about 5.4% loss at six months. Noom's own largest trial reported 4.1% at 68 weeks against a 1.5% gain in controls, though the in-program 16-week difference was only 1.8 percentage points and the trial was company-funded and company-announced.

Fasting protocols occupy the middle tier, and the tier is only visually distinct. The rates look higher because the protocols are more aggressive, not because the mechanism is different. OMAD sits at the top of the fasting range for the unglamorous reason that eating maintenance calories in one sitting is difficult. It also has the thinnest evidence base of any protocol here, an LDL-increase signal, and the worst structural problem with protein.

Supplements occupy the bottom tier and it is a long way down. Green tea catechins with caffeine, the best-evidenced entry in the whole aisle, moved 1.31 kg over 12-13 weeks against control. Berberine moved 0.88 kg across 23 trials, and that effect does not survive the exclusion of low-quality studies. Synephrine has no demonstrated weight-loss effect at all and does raise blood pressure and heart rate.

Adherence explains more of the variance than method choice

In an app-based commercial trial, people who logged at least six days a week in at least 75% of weeks lost more than inconsistent loggers at one, three and six months. Same app, same food database, same features. The variable that moved the outcome was whether the person kept doing it.

This reframes the choice you are actually making. You are not picking the most effective method, because within a tier they are close to interchangeable. You are picking the one whose failure mode you can live with. Calorie counting fails when you get bored of measuring. 5:2 fails when a fast day collides with a birthday dinner. 16:8 fails least often, which is part of why its measured effect is modest: it asks little and it delivers proportionally.

Two structural facts help more than any protocol. Rates above roughly 1% of body weight per week increasingly take lean mass with them, so aggression has a ceiling that is not about willpower. And protein plus resistance training barely register on the weekly rate while doing most of the work on what the lost weight is made of.

Who each approach actually suits

Match the method to how you fail, not to how you hope to succeed.

• You want structure and data, and you will use it: calorie counting or MyFitnessPal. Highest ceiling, highest measurement burden.
• You want structure but counting feels punishing: WW Points. A weighted calorie proxy that nudges food choice as well as quantity.
• You quit for motivational reasons rather than informational ones: a coached program like Noom, with the caveat that its measured in-program effect is small.
• You would rather change when you eat than what you eat: 16:8 first. It has the lowest dropout of any fasting protocol and the smallest ask.
• You would rather be strict occasionally than moderate constantly: 5:2, or alternate-day fasting if you can tolerate the dropout risk.
• You have blood glucose concerns alongside weight: an energy-reduced Mediterranean pattern has the best combined evidence for visceral fat and adherence.
• You are already lean-ish and worried about muscle: protein at 1.2-1.6 g/kg/day plus two to four resistance sessions, with a slower rate target.

What none of these do

None of them suppress appetite the way a GLP-1 agonist does. That is the entire mechanism gap. Fibre raises endogenous GLP-1 through short-chain fatty acids, which is a genuine mechanism operating at a fraction of pharmacological magnitude. Berberine does not act on the GLP-1 receptor at all, whatever TikTok called it.

None of them reliably hold the result without ongoing effort. That is also true of the drugs, as the STEP 1 extension showed, so it is not a criticism unique to the natural side.

If you have a BMI and comorbidity profile where a GLP-1 is clinically indicated, the honest answer is that the methods on this page are unlikely to reach the same endpoint, and a conversation with a prescriber is the right next step. Using this section as a reason to avoid that conversation is the one way to read it wrong.

If you are starting today

Pick one deficit method and one composition method, and give it eight weeks before judging it. One from the top of the table, one from the protein and training rows. Anything from the supplement tier is optional and, on the published numbers, roughly decorative.

• Set a rate target between 0.25% and 1.0% of body weight per week. The CDC's 1-2 lb per week guidance lands in that band for most adults, and gradual loss is associated with better maintenance.
• Choose the deficit method whose failure mode you can tolerate. Do not choose two from the same row of the redundancy table.
• Set protein at 1.2-1.6 g/kg/day and add two resistance sessions per week. This barely moves the scale and substantially changes what comes off.
• Protect sleep. In the Nedeltcheva crossover, short sleep did not change total weight lost, it cut the fat fraction of that loss by 55%.
• Weigh consistently and read the trend, not the day. Weekly noise routinely exceeds a week of genuine progress at rates below 0.25%.

The evidence, one row per claim

Questions people ask

Sources

1. BMJ Group (network meta-analysis summary)
2. Wilding et al., Diabetes Obes Metab (STEP 1 extension)
3. CDC — Steps for Losing Weight
4. Systematic review of control-group weight loss in lifestyle RCTs, PMC

Keep reading

• How Fast Should You Actually Lose Weight?Numeric and decision-shaped. A floor, a ceiling, and the reason both exist.
• Energy Balance: The One Mechanism Behind Every Method on This SiteThe mechanism page. Every other page here is a different way of arriving at the same physiology.
• WeightWatchers Points: How the System Works and What the Trials ShowPoints explained as calorie counting with deliberate distortions, plus the real trial numbers and the GLP-1 pi
• Intermittent Fasting for Weight Loss: What 99 Trials SayHub for the fasting cluster. Fasting is a scheduling tool for calorie restriction, not a separate metabolic le
• Caffeine, Green Tea and Thermogenic Fat Burners: Small Effects, Honestly ReportedCategory page for fat burners, with effect sizes reported at their real size including the zeros.
• Coming Off a GLP-1: Building the Habits That Hold the LossThe highest-stakes page here. Regain is the default outcome, the numbers are known, and the behavioural side h