How Accurate Are Body Fat Scales? An Honest Guide to BIA

Body fat scales are not lab-accurate and we explain exactly why. Error margins, what BIA can and cannot measure, and how to get readings you can trust.

Body fat scales are not laboratory instruments, and any brand that tells you otherwise is overselling. Consumer BIA scales typically read body fat within 3 to 10 percentage points of a clinical DEXA scan. What they do reliably is repeat the same measurement the same way, which makes them genuinely useful for tracking change over weeks. Here is the honest picture.

What a body fat scale actually measures

Step on the platform and the scale sends a tiny, imperceptible electrical current through your body. Muscle and other lean tissue hold a lot of water, so they conduct electricity well. Fat tissue holds very little water, so it resists the current. The scale measures that resistance — the impedance — and feeds it into an equation along with your height, age, sex and weight. The output is an estimate of your body fat percentage.

That word, estimate, is doing a lot of work. The scale never counts fat cells. It measures one electrical property and infers everything else from population statistics. Two different brands will use two different equations, which is why your reading can jump several points when you switch scales, even on the same morning. Nothing about your body changed; the algorithm did.

The honest error numbers

Independent research gives us a reasonable range rather than a single figure:

• Consumer BIA devices generally land within 3 to 10 percentage points of DEXA, the clinical reference method used in hospital body composition research.
• Devices that measure through hands and feet together — the 8-electrode design — narrow that band to roughly 3 to 5 points in academic testing. Foot-only scales, the common 4-electrode design, sit closer to 4 to 8 points.
• A study in BMJ Nutrition, Prevention & Health found a standing BIA system tracked DEXA closely but with a systematic offset of about 3 percentage points. The offset was stable, which matters more than it sounds: a consistent error still shows you the true direction of change.
• A 2022 validation study found within-session repeatability was excellent (coefficient of variation under 2%) across three BIA devices, even though absolute agreement with DEXA was poor at the individual level.

One independent lab test of an Arboleaf model (the CS20N, a foot-only design) measured weight accurate to ±0.18 kg and body fat reading about 4.5 points above the DXA reference. We cite that number rather than hide it, because it is exactly what the science predicts for a 4-electrode scale — and it is why the model sold on this site uses an 8-electrode, dual-frequency design instead. You can read how the two architectures differ in our 8-electrode vs 4-electrode comparison.

Why your reading can look wrong

If your scale says 52% one month and 38% the next when the mirror barely changed, you have not lost 14 points of fat. You have seen the equation react to changed inputs. The main culprits:

Hydration. BIA is, at its core, a water measurement. Weigh in dehydrated after a workout or a night of alcohol and your fat reading climbs, because less water looks like less lean tissue to the current. Drink two pints and it falls.

Time of day. Overnight, water redistributes and you lose weight through respiration. Morning and evening readings on the same scale can differ by more than a kilogram of weight and several points of body fat. Pick one slot and keep it — we explain the mechanism in our morning vs evening weigh-in guide.

Food and glycogen. Every gram of stored carbohydrate binds roughly three grams of water. A high-carbohydrate day pulls water into muscle and can make you read leaner; a low-carbohydrate day does the reverse.

Feet and surface. Dry, hard skin on your soles raises impedance. Carpet under the scale flexes the platform and distorts the load cells. Bare feet, hard floor, every time.

The equation itself. Population equations are least reliable for people far from the population average — very lean athletes, petite frames, older adults. This is where segmental 8-electrode measurement earns its keep, because it measures limbs and trunk separately rather than assuming a standard body layout.

So are they worth using at all?

Yes — if you use them for the thing they are actually good at. The research consensus and the honest reading of the numbers point the same way:

• As a single absolute number: treat with caution. Do not compare your 24% against a friend’s 21% from a different brand. Do not compare either against a DEXA printout.
• As a trend line: genuinely useful. Because the error is largely systematic and repeatability is high, week-over-week direction is real information. If your four-week average is falling while weight holds steady, something meaningful is happening to your composition.
• As motivation infrastructure: quietly powerful. Automatic logging removes the friction and the selective memory from tracking. The reading happens, the app records it, the trend builds.

That is the entire philosophy behind how we built this product page and this blog: the Arboleaf 8-electrode scale is sold as a trend instrument, not a diagnostic device. It does not diagnose anything, and no bathroom scale should claim to.

How to get readings you can actually trust

Follow a consistent protocol and BIA becomes far more dependable:

1. Weigh at the same time of day — first thing in the morning, after the loo, before food or drink is the easiest slot to standardise.
2. Bare feet on a hard, level floor. Never carpet.
3. Same conditions: avoid weighing straight after exercise, a sauna, or a heavy meal.
4. Look at the weekly average, not the daily number. Day-to-day noise of one to two kilograms is normal water movement, not fat.
5. Give any new scale two to four weeks of data before judging your trend. One reading tells you almost nothing; twenty readings tell you a lot.

We keep a full measurement-conditions checklist in our FAQ, and if a reading ever seems implausible, re-weigh once under the same conditions rather than chasing the number around your bathroom.

The bottom line

How accurate are body fat scales? Accurate enough to show you where you are heading, not accurate enough to tell you exactly where you stand. The absolute number is a reference point with a known error band; the trend is the product. Any scale marketed as clinically precise deserves your scepticism — and a scale that is honest about its limits, measured consistently, will still tell you more about your progress than the mirror, the tape and guesswork combined.

If daily numbers affect your mood more than they inform your decisions, weigh weekly instead. The data serves you, not the other way round.