How Accurate Are Body Fat Scales?

A home body-fat scale reads your weight to about a tenth of a kilo and estimates your body fat to within roughly 3–5 percentage points of a DEXA scan. That's too wide to treat a single reading as a fact and narrow enough—because the error is consistent for the same person under the same conditions—to make the trend over weeks genuinely useful.

What the scale is actually doing

A tiny alternating current passes up one leg and down the other. Fat resists it more than muscle and water do, so the resistance (impedance) implies a tissue mix. But the scale can't see tissue—it puts that impedance figure into an equation along with your height, age, and sex and returns an estimate. Two things follow from that, and they explain almost every complaint about these scales.

  1. The current takes the shortest path—up your legs and back down. It barely samples your torso and arms, so a foot-to-foot scale infers your upper body rather than measuring it.
  2. The equation was built on a reference population. The further you are from that population—very lean, very muscular, elderly, pregnant, or an amputee—the further the estimate drifts.

What moves the number without your body changing?

Factor

Effect on the reading

Why

Hydration

Up to several points 

Water conducts. Dehydrated reads as higher body fat; over-hydrated as lower. 

Time of day 

1–2 points 

Fluid redistributes overnight; morning readings are the most consistent. 

A salty or large meal 

1–2 points 

Shifts body water and gut contents. 

Hard training in the last few hours 

1–3 points 

Sweat loss plus blood redistribution. 

Alcohol the night before 

1–3 points 

Diuretic effect. 

Damp or sweaty feet 

Falsely low 

Better conduction at the electrode. 

Socks or dry, calloused soles 

No reading, or a wild one 

Poor contact. 

Menstrual cycle phase 

1–2 points 

Cyclical fluid retention. 

 

What it's good at

  1. Weight: ±0.1 kg on four load cells. Not an estimate.
  2. Direction of change over weeks. The error is largely systematic, so a consistent downward or upward trend in the estimate is real even if the level isn't.
  3. Consistency of conditions. Same scale, same time, same state—comparable readings.
  4. Cost. A DEXA scan costs more per scan than the scale costs once.

What it's bad at

  1. Absolute accuracy. 27% on your scale means somewhere around 22–32% in truth.
  2. Comparing across scales. Two brands use different equations and will disagree by several points on the same person on the same morning.
  3. Comparing between people. Your 22% and your partner's 22% aren't the same measurement.
  4. Day-to-day change. That's hydration.
  5. Edge populations. Very lean, very muscular, pregnant, elderly, or anyone with an implanted electronic device—the last group should not use bioimpedance at all.

How to use one so it's worth having

  1. Weigh once a week, same day, same time—first thing, after the bathroom, before eating or drinking.
  2. Bare feet, dry soles, stand still until the reading settles.

  3. Record every reading; never delete a bad one.

  4. Read the seven-day average, and compare month to month. 
  5. Ignore any single reading that surprises you. Take next week's. 

The honest summary 

If you want to know your body-fat percentage as a fact, a home scale is the wrong instrument—book a DEXA scan. If you want to know whether the direction of travel over three months is what you expected, a £40 scale used consistently will tell you that about as well as a £4,000 machine used occasionally. Those are genuinely different questions, and the scale industry makes its money by blurring them. 

Frequently asked questions 

Q. How accurate are body fat scales?

A. Within roughly 3–5 percentage points of a DEXA scan for a single reading. Weight itself is accurate to about ±0.1 kg. Because the error is consistent for the same person under identical conditions, the weekly trend is far more trustworthy than any individual figure.

Q. Why do two scales give me different body fat percentages?

A. Because each manufacturer uses a different estimation equation on the same impedance measurement. Two scales can differ by several points on the same person on the same morning, and neither is necessarily wrong. Never compare readings across devices.

Q. What's the most accurate way to measure body fat at home?

A. None of the home methods is accurate in absolute terms. The most useful approach is consistency: one scale, same time of day, same conditions, weekly, reading the trend. For an absolute figure, a DEXA scan is the practical reference standard.

Q. Do body fat scales work if you're very muscular or very lean?

A. Less well. The estimation equations are built on general-population data, so readings drift at the extremes of body composition. The trend remains useful; the absolute number should be treated with more skepticism than usual.

Q. Should I stop using my smart scale if it's not accurate?

A. Not if you use it for what it's good at. Weight is a real measurement, and the composition trend over weeks is informative. It's the single-reading body-fat number that shouldn't be treated as a fact.  

Back to blog