A smart scale's muscle mass is a brand's own compartment, not the muscle you can flex. One published equation, Janssen 2000, estimated MRI-measured skeletal muscle with a 2.7 kg (9 percent) standard error, fitted in 269 white adults on a lying-down hand-and-foot analyser. It tested no scale.
Why a muscle number can mislead
Tanita makes body-composition scales, and its support page answers the question "why is my result for total muscle mass so high?" Its answer: the figure counts skeletal muscle, smooth muscle and organs such as the digestive organs, lungs and heart, "together with all the water they contain" (Tanita).
Our view: a label that says "muscle" for that compartment invites the wrong reading. Many readers will see the word and think of the muscle they train. The scale is reporting something wider, so this post says "lean" for what a scale estimates.
Two terms the scale blurs
Fat-free mass is a research term with a plain definition. The ESPEN guideline on bioelectrical impedance analysis says fat-free mass is everything that is not body fat (ESPEN). Bone, organs, blood, water and muscle all sit inside it.
Skeletal muscle is one tissue inside that total: the striated muscle you control voluntarily, most of it attached to bones through tendons, as Renpho describes it. Tanita's and Renpho's "muscle mass" figures both sit between those two definitions, and the label alone does not say where.
What the makers say
Here are the definitions on the makers' own pages, read on 2026-10-20. "Not stated" means the page does not say.
Table: what each source says its figure includes, read 2026-10-20. Tanita and Renpho are makers; ESPEN is a clinical guideline.
| Source and term | Water | Organs | How the page describes it |
|---|---|---|---|
| Tanita, total muscle mass | Included, per the page | Included, per the page | Skeletal muscle, smooth muscle and organs such as digestive organs, lungs and heart |
| Renpho, muscle mass | Included, per the page | Not stated | Smooth and skeletal muscle, plus water in the body |
| Renpho, skeletal muscle | Not stated | Not stated | Striated muscle under voluntary control; no percentage definition stated |
| Renpho, fat-free mass | Included, per the page | Included, per the page | Bone, muscle, connective tissue, brain, skin, organs, blood, water and stored chemicals |
| ESPEN guideline, fat-free mass | Included, about 73 percent in healthy adults | Included | Everything that is not body fat |
Renpho's page lists muscle mass, skeletal muscle and fat-free mass as separate entries, and it gives no denominator for a skeletal muscle percentage. That is why "muscle mass 62 kg" beside "skeletal muscle 38 percent" cannot be read as one number and its share. This is an illustrative reading, not a measured case.
What the validation shows
Janssen 2000 is the validation paper this post leans on. The authors measured 388 healthy adults, aged 18 to 86, against MRI, and built the final equation in 269 of them. It used one lab analyser with electrodes on the hand and foot while people lay down (Janssen 2000). A standing scale with foot plates is a different device.
For scales, one study stands out. Frija-Masson 2021 tested three connected scales against DXA in hospital patients, each patient on one scale, measured once. They were adults referred for a body-composition scan, many with obesity or chronic conditions. The Results section reports median muscle errors of +4.5 kg for the Tefal Body Partner (53 patients), -6.6 kg for the Terraillon DietPack (52) and +4.0 kg for the Withings Body Cardio (48). The abstract prints different figures, so check the paper before quoting either set.
Those results belong to those three models in those patients. They say nothing about a scale released since. The limit cuts both ways: these two studies do not show that your scale is accurate, and they do not show that it is not.
Why two scales disagree
Two things differ between devices. The compartment differs between Tanita's and Renpho's definitions, as the table shows. The equation can differ too, and the authors of the 2021 study say the scales' regression equations are unknown (Frija-Masson 2021). The ESPEN guideline reports that average fat-free mass is about 73 percent water in healthy adults, and that equations developed under normal hydration have not been shown valid when hydration is altered.
So a rising figure is not proof of muscle gained. It can reflect a shift in the water the equation assumes. In the 2021 study, the median errors of the three scales had different signs, in three different groups of patients. A 4 kg gap between your scale and a friend's is not evidence that either one is right. Say one scale shows 62 kg of muscle mass and another shows 58 on the same morning (an illustrative pair, not a measured case). The honest reading is that two makers drew the compartment or the equation differently, not that you lost 4 kg of anything.
How to read your own line
Check your app's label. If the definition is not on the maker's page, treat the figure as unexplained.
Our position: the kilogram figure is the weakest line on the screen, because it carries the most definition and the least context. An estimated lean line tracked against itself tells you more than a muscle figure compared with someone else's.
Then watch one line, on one scale, under fixed conditions, over weeks. Day-to-day movement is noise in the estimate. If you need to know whether you gained muscle, a scale is the wrong instrument: the studies above used MRI and DXA as their references, and a scan is a question for a clinician.
A change in scale weight or body-fat estimate does not establish exactly how much fat, muscle, water, or other tissue changed. Trendline avoids making that claim. The same logic is in our guide on why consistency beats accuracy for body-fat scales.
If watching any of these numbers starts to drive your eating or your mood, it is fine to stop and talk to a doctor.
Lean estimate, labeled
We make Trendline. See weight, body fat, and estimated lean mass together in it, with device estimates labeled as device estimates. Trendline reads compatible measurements through the Apple Health app, not directly from every scale.
Questions people ask
Is the muscle mass on my smart scale real muscle?
Not necessarily. Tanita's support page says its total muscle mass includes organs and the water they contain. Treat the figure as the maker's estimate of a defined compartment, and read your app's own label.
What is the difference between lean mass and muscle mass?
Lean or fat-free mass is everything that is not body fat, which includes bone, organs, water and muscle. Skeletal muscle is one tissue inside it. A scale's muscle mass is the maker's own compartment, and it can include more than skeletal muscle.
Why do two scales show different muscle mass on the same morning?
Tanita's and Renpho's definitions differ, and the authors of one 2021 hospital study say the scales' regression equations are unknown. In that study, the median muscle errors of three scales had different signs. Compare a figure only with the same device.
Does a rising muscle number on my scale mean I gained muscle?
No. A scale estimates a compartment that can include water and organs, and the ESPEN guideline says equations developed under normal hydration have not been shown valid when hydration is altered. A rising figure is a reason to look at strength, tape measurements or a clinic scan, not proof of new muscle.
Which line on my scale app should I watch?
Pick one line, such as estimated lean mass, and watch it over weeks on one scale under fixed conditions. Treat day-to-day moves as noise in the estimate, and do not use the kilogram figure to decide whether you gained muscle.