How Fast Should You Actually Lose Weight?

The working range on this site is 0.25% to 1.0% of body weight per week. The ceiling exists because loss above roughly 1% a week increasingly comes out of lean tissue. The floor is not a safety limit at all, it is a practicality limit: below 0.25% a week, normal weight fluctuation is larger than your progress and most people cannot tell whether the plan is working. Everything else on this page is about why those bounds are expressed as a percentage, and why the first two weeks will mislead you regardless of what you do.

The standard guidance and where it comes from

The CDC recommends 1 to 2 lb, roughly 0.45 to 0.9 kg, per week, and notes that people who lose weight at that gradual pace are more likely to keep it off. Harvard Health frames a healthy realistic rate in similar terms and explains why the rate itself matters for maintenance. NIH consumer guidance reinforces gradual weight control as the approach associated with durable results.

That is three national-level sources landing in the same place, which is about as settled as this field gets. The problem is the unit.

Why percent of body weight beats pounds per week

One to two pounds a week means completely different things to two different people, and the guidance does not distinguish between them. For a 90 kg adult, 1 to 2 lb is roughly 0.5 to 1.0% of body weight a week. For a 65 kg adult, the identical advice is 0.7 to 1.4% a week, which pushes past the point where lean-mass loss climbs.

The same absolute deficit is a mild deficit for a larger body and an aggressive one for a smaller body, because a larger body has both a higher total expenditure and more fat available to draw from. Expressing the target as a percentage scales automatically and stops a smaller person from following advice that was calibrated for someone else.

It also self-corrects as you go. A 1% target at 100 kg is 1 kg a week. At 80 kg the same target is 800 g. The rate slows on its own, which is the correct behaviour and something a fixed pounds-per-week target never does.

The first two weeks lie

Almost everyone loses 2 to 4 kg in the first fortnight of a diet, and it is mostly not fat. Cutting calories, and especially cutting carbohydrate, depletes muscle and liver glycogen, and each gram of stored glycogen holds roughly three grams of water. Reduced sodium and food volume in the gut contribute too.

It comes back the moment carbohydrate intake rises, which produces the demoralising experience of gaining 2 kg over a weekend and concluding the whole effort has failed. Nothing failed. Water moved.

Judge your actual rate from about week three onward, using a rolling average across three or four weeks rather than any single weigh-in.

The ceiling: what goes wrong above 1%

Lean mass is the main cost. The faster the loss, the greater the proportion that comes from muscle rather than fat, and the muscle does not come back on its own when you stop dieting. That is the primary reason for the ceiling.

Composition is not decided by rate alone. Protein intakes of 1.2 to 1.6 g/kg/day attenuate lean-mass loss under energy restriction, and resistance training does the rest. Sleep matters as much: in a crossover trial on identical calorie restriction, sleep restriction cut the proportion of weight lost as fat by 55%. You can run near the ceiling with protein, lifting and sleep covered. Running near the ceiling without them is where the damage happens.

Very steep deficits also make nutrient adequacy harder, since there is simply less food to carry your micronutrients, and rapid weight loss is associated with increased gallstone risk. Both are reasons to prefer the middle of the range.

The floor: when slow is too slow

The 0.25% floor is not a safety threshold. It is the point below which the signal disappears into the noise. Day-to-day body weight swings by a kilogram or more from water, food volume and hormonal cycles, so a 200 g weekly change is invisible for weeks at a time.

That invisibility kills adherence. If you cannot see whether the plan is working, you cannot tell whether to adjust it, and most people quit rather than continue on faith. A slower rate is a legitimate choice for someone lean or someone prioritising training, and below the floor it stops being a plan you can steer.

Higher starting weight, faster safe rate

Someone at 130 kg can sustain a larger absolute deficit than someone at 70 kg, because total daily expenditure is higher and fat mass supplies more of the energy. Using percentages handles this automatically, which is the whole argument for the unit.

As you get lighter, move down the band. Starting at 1.0% at 120 kg and finishing the last few kilograms at 0.4% is a sensible progression, and it matches the fact that the final stretch is where lean-mass risk is highest and the deficit is hardest to find.

Rapid loss on a GLP-1

The rules do not change, only the cause does. Semaglutide 2.4 mg produced about 15 to 17% body weight loss at 68 weeks in trial conditions, and the early phase for some people runs above 1% a week. That is faster than anything this page recommends achieving by diet alone.

The response is not to slow the medication on your own. It is to cover the things that decide composition at that rate: protein at the upper end of the range, resistance training, and sleep. Any change to a dose, a titration schedule or whether to continue treatment is a decision for your prescriber. If you want to understand the arithmetic of a given schedule before that appointment, our /tools/microdose-calculator estimates it and displays the assumptions behind the estimate rather than just returning a number.

A simple target-setting method

Pick a percentage, convert it to grams per week at your current weight, recalculate every 5 kg, and measure the trend over a month rather than a week.

Most people should sit between 0.5% and 0.75% a week. That is fast enough to see, slow enough to hold, and it leaves headroom for a bad fortnight without abandoning the whole thing.

• Choose a percentage between 0.25% and 1.0%, nearer the middle than the top
• Multiply by current body weight for a weekly target in grams
• Ignore the first two weeks entirely
• Use a rolling three to four week average, not single weigh-ins
• Recalculate the target every 5 kg lost

The evidence, one row per claim

Questions people ask

Sources

1. CDC — Steps for Losing Weight
2. Harvard Health
3. NIH News in Health
4. Nedeltcheva et al., Annals of Internal Medicine
5. Clin Nutr ESPEN

Keep reading

• 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.
• Protein and Lifting: How to Lose Fat Without Losing MuscleThe most actionable page here. A deficit decides how much you lose. Protein and lifting decide what.
• Alternate-Day Fasting and Extended Fasts: The Strongest Signal and the Biggest RisksTwo subjects on one page: the protocol with the best comparative data, and the practice with real medical risk
• Why Weight Loss Stalls: Metabolic Adaptation, Honestly AssessedTwo research camps, both cited. The phenomenon is measurable, the excuse built on it is not supported.
• Sleep and Alcohol: The Two Variables That Change What You LoseNeither changes the scale much. Both change what the scale is made of, which is why people dismiss them.