How Often Should You Weigh Yourself?

Weigh as often as you will actually average it#

The frequency that fits the evidence is daily or near-daily — but not because weighing more often makes you lose more. When researchers randomly assigned 183 obese adults either to receive scales with instructions to weigh and record daily, or to a control condition, the daily-weighing group was 0.5 kg lighter at three months, with a confidence interval running from −1.3 kg to +0.3 kg (p = 0.24). Within the intervention group, each additional day of weighing was worth 20 g of extra loss, 95% CI −30 to +20 g¹. The authors' verdict is one sentence long:

Advice to weigh oneself daily is ineffective as a sole strategy.

So why does daily weighing have the reputation it has? Because almost all the encouraging evidence comes from a different kind of study — one that watched who chose to weigh often, inside a program that was already doing the work. Both literatures are right, and the space between them is the real answer to the question in the title: extra weigh-ins buy resolution, not fat loss. They give you enough readings to average, and an average is the only form in which your weight reports anything about the deficit you are running. Read as a morning verdict, the same number is a mood.

Assignment and association point in opposite directions#

This is not a case of good studies disagreeing about a fact. It is one question asked two ways, and the wording of the question decides the answer.

The association side is genuinely strong. A systematic review of 17 longitudinal studies in adults seeking weight-loss treatment found regular self-weighing consistently associated with more weight lost, and not with depression or anxiety². The most-quoted single result — daily weighers losing about 6 kg more than less-frequent weighers over six months — is covered in the habits article, and two details of its design belong here rather than there: 47 people, and the "less frequent" comparison group still weighed itself an average of 5.4 days a week⁶. That is a contrast between weighing seven days and weighing five, in a sample the size of a school class, with nobody assigned to anything.

The largest objective dataset points the same way and puts a size on it. Across 9,768 smart-scale users, 4.2 million weigh-ins and a mean follow-up of nearly three years, the correlation between how often someone weighed and how much weight they changed was r = −0.111³. Real, highly significant, and about one percent of the variance in weight change — that squared figure is my arithmetic on their coefficient, not a number the paper prints. Frequency is a signal in the noise, not the engine.

What separates the two literatures is not sample size or quality. It is that weighing frequency is mostly downstream of engagement: the person who steps on the scale every morning is the person still running the plan, and the plan is what moves the weight. Randomization cuts that link, and when it is cut the effect largely goes with it.

What the extra readings do buy#

Stop asking the scale to be a treatment and the case for frequency becomes clear and much narrower.

The first thing more readings buy is a usable average. Day-to-day body mass swings by more than a good week of fat loss, for reasons that have nothing to do with fat — glycogen water, gut contents, a hormonal phase — so a single morning is one draw from a wide distribution, as the physiology of the daily swing sets out. Averaging is the only fix, and you cannot average readings you did not take. How many days it takes before a flat line means something is worked out in plateau or fluctuation; the relevant point here is that the arithmetic assumes you have the days.

The second is less obvious and, in the smart-scale cohort, more dramatic than any frequency effect: what happens when people stop. Of those 9,768 users, 72.5% took at least one break of 30 days or longer, with a median break of 59 days — and the median weight change across a break was a gain of 0.85 kg, rising to 1.37 kg in users with obesity, with longer breaks tracking larger gains³. Read that as correlation, because it is one: people who stop weighing are usually people who have stopped doing everything else too. But it does reframe the practical question. The gap that matters is not daily versus weekly. It is weekly versus the two months where you quietly stopped looking — which is why the maintenance literature cares about whether you are still looking at all, rather than how often.

Where the evidence genuinely splits: who is holding the scale#

On psychological harm, adult treatment populations look clean. Zheng's 17 studies found no association with depression or anxiety, and a six-month randomized trial in 91 overweight adults assigned to daily weighing found no penalty on any of five psychological measures, with body dissatisfaction actually falling further in the weighing arm (p = 0.007)⁴ — the full outcome list, and the trial that complicates it, sit in reviewing calories daily or weekly.

A different population gives a different answer. Project EAT followed 1,868 people from adolescence into young adulthood, measuring three times across ten years. Among the 1,058 women, increases in self-weighing tracked increases in weight concern (effect size 0.21) and depressive symptoms (0.09), and decreases in body satisfaction (−0.11) and self-esteem (−0.10), plus higher odds of extreme unhealthy weight-control behaviors (OR 1.21). Among the 810 men, only weight concern moved (0.24). In neither sex did self-weighing predict any change in BMI⁵.

These two results are not in conflict, and the thing that separates them is nameable. One is a randomized six-month intervention in adults who came looking for weight-loss treatment, where the scale arrives attached to a plan and a purpose. The other is an observational cohort of teenagers weighing themselves for their own reasons, where nothing establishes that the weighing caused the distress rather than the distress causing the weighing. The effect sizes on the harm side are also small — 0.09 to 0.24 — and no benefit appeared to offset them.

One detail deserves to be stared at. The adult trial recorded a significant rise in dietary restraint in the daily-weighing arm and reported it as a good outcome; the adolescent cohort recorded a rise in weight concern and reported it as a bad one. Those are close cousins measured on different instruments in different lives. The behaviour is the same; what changes is whether it is pointed at something. If the daily number is doing nothing except making you feel watched, that is a reason to weigh less often, not evidence that you lack discipline — and the wider question of when self-monitoring tips into something worse is handled in tracking without obsessing.

A protocol that makes the readings comparable#

Because the value of frequency is resolution, everything in the protocol exists to stop consecutive readings from differing for reasons other than you. Research protocols that need day-to-day comparability converge on the same conditions: participants weigh themselves each morning, unclothed, fasted, after voiding, on the same scale⁷. There is no true weight that this finds. There is a repeatable weight, which is the only kind an average can use.

Decision	Do this	Why
How often	Daily, or a fixed weekday each week	Enough draws to average; a fixed weekday removes the weekly rhythm
When	Morning, after voiding, before eating or drinking	The most reproducible point in the day
What you read	A 7-day rolling average, judged over 3+ weeks	A single morning is noise; the trend is the fat signal
What you act on	A pre-set trip line, not today's number	Turns a reading into one decision instead of a daily verdict
When to weigh less	If the number is running your mood and nothing else	The harm signal is affective, not numerical

And one thing not to do: never compare today against yesterday. The comparison the evidence supports is this fortnight's average against the last, which is a question the scale can answer, rather than "am I up?", which it cannot.

FAQ#

Is daily weighing better than weekly for losing weight?#

Not on its own. Randomly assigning adults to weigh and record daily produced a 0.5 kg difference at three months with a confidence interval spanning zero, and within that trial each extra weighing day was worth 20 g¹. Daily weighing is strongly associated with more loss in observational data², but that mostly reflects who chooses to do it. Weigh daily for a better-resolved trend; do not expect the weighing itself to be the intervention.

When is the best time of day to weigh yourself?#

In the morning, after using the bathroom, before eating or drinking, on the same scale — the conditions research protocols standardize on when they need day-to-day comparability⁷. The point is not that morning weight is your real weight. It is the most reproducible moment in the day, so consecutive readings differ because of you rather than because of a lunch.

Can weighing yourself too often be harmful?#

In adults in weight-loss treatment, trials have not found it so: a six-month randomized trial found no increase in depressive symptoms, binge eating or disinhibition⁴. In a ten-year cohort of adolescents and young adults, rising self-weighing tracked rising weight concern and, in women, more depressive symptoms and more extreme weight-control behaviors, with no BMI benefit⁵ — though the effect sizes were small and the design cannot show direction. The practical read: a scale attached to a plan is a tool, and a scale attached to nothing is a mirror.

Sources#

This article was researched and drafted with AI assistance and reviewed for accuracy by the BurnWeek team. It is general information, not medical advice. How we research and correct our articles →