Start With What You Already Own
Most advice on this topic answers a question nobody asked: which body fat method is best in theory. The DEXA scan wins that argument every time, and it helps almost nobody standing in their bathroom on a Tuesday morning wanting a number.
The better question is which method is best given your actual constraints — what equipment is within arm's reach, how much friction you will tolerate, and how much you are willing to spend. Answer that honestly and you will pick a method you can repeat for six months, which matters far more than the theoretical accuracy of a method you use once.
Three questions route you to the right branch:
- Do you have a flexible tape measure, or can you get one this week?
- Is there a body composition scale in your house, or a gym or clinic you already visit?
- Do you have a budget between $50 and $150 for a one-time scan?
| Your situation | Recommended method | Typical accuracy | Cost |
|---|---|---|---|
| Phone only, nothing else | Photo AI estimate | ±3–5 points | Free |
| Tape measure available | Navy circumference formula | ±3–4 points | $6 tape |
| Smart scale at home | Scale as daily trend, tape as anchor | ±5–8 points alone | Already owned |
| Gym or clinic access | InBody or trained caliper test | ±3–5 points | Often included |
| $50–150 to spend once | DEXA or BodPod for calibration | ±1–3 points | $50–150 |
Each branch below gives the protocol, the honest error range, and what to pair it with. All of these are directional estimates for tracking your own progress, not medical diagnosis.
Branch A: Phone Only, No Equipment At All
This is the most common starting point and the least well served by traditional advice, which usually just tells you to buy something.
What to do. Use a photo-based estimate. The photo AI estimator works from a front and a side image and needs nothing but a phone and a wall. Wear form-fitting clothing or swimwear so contours are visible, stand six to eight feet from the camera with the lens at roughly chest height, keep your arms slightly away from your sides, and do not brace or suck in. Take the side shot from exactly the same spot after a 90-degree turn.
Expected accuracy. Around ±3–5 percentage points against a reference scan, which puts it in the same broad class as tape methods. Repeatability is the bigger variable: with a fixed setup you can hold session-to-session scatter near 1 point, but if lighting, distance, or clothing drift between check-ins that scatter can double.
What makes or breaks it. Lighting is the single largest controllable factor. Overhead light casts shadows that carve definition where none exists; a window in front of you at mid-morning gives flat, honest illumination. Take one reference screenshot of your setup — where you stand, where the phone sits, what you wear — and copy it literally every time.
Free additions that cost nothing. A piece of string and a ruler will give you a waist measurement even without a tape: wrap, mark, lay flat, measure. Log that number alongside your estimate. Waist circumference is the single most informative raw measurement you can collect, and it needs no equation at all to be useful.
When to leave this branch. As soon as you can spend six dollars. A photo estimate plus a tape estimate is meaningfully better than either alone.
Branch B: You Have a Tape Measure
A flexible cloth or plastic sewing tape costs about $6 and unlocks the best-validated home method available.
What to do. Run the Navy circumference formula through the Navy body fat calculator. Men need height, neck, and waist. Women need height, neck, waist, and hip. Measure first thing in the morning, before eating, with the abdomen relaxed.
Landmarks that matter. Neck: just below the larynx, tape sloping slightly downward at the front. Waist for men: at the navel. Waist for women: at the narrowest natural point. Hip: at the widest part of the glutes, feet together, tape level all the way around. Snug enough to stay in place, not tight enough to compress the skin.
Expected accuracy. ±3–4 percentage points for most adults, with two known failure modes. Heavily muscled lifters often read high because thick obliques and a developed midsection inflate the waist term. People with a slim frame but significant abdominal storage can also read high. Both biases are consistent for you personally, which means they cancel when you compare two of your own readings.
Why duplicates matter more than technique perfectionism. Half an inch of waist error moves the Navy result by roughly 0.9 percentage points — often larger than a month of real progress. Measure each site twice and average; take a third if the two differ by more than half an inch. This one habit does more for your data quality than anything else in this article.
Pair it with. A photo estimate monthly. The two methods fail differently, so agreement between them is genuinely informative. If your concern is abdominal fat specifically rather than total composition, the visceral fat calculator uses waist and height to screen that directly.
Branch C: There Is a Smart Scale in the Bathroom
Bioimpedance scales are the most owned and least understood tool in this list. They are not useless, but they answer a narrower question than their display suggests.
How they work and why that limits them. A small current passes up one leg and down the other, and the scale infers composition from electrical resistance. Fat conducts poorly, lean tissue conducts well because it holds water. The problem is that hydration status changes resistance far more than a week of fat loss does. Foot-to-foot models also sample mainly your lower body and estimate the rest.
Expected accuracy. ±5–8 percentage points against DEXA, with a known tendency to read low for people carrying more fat and high for lean, muscular users. Same-day scatter can reach 2 points from nothing but a glass of water and a walk.
How to actually use one. Treat the absolute number as unreliable and the daily series as a smoothing tool. Step on at the same time every morning, after using the bathroom, before drinking anything, barefoot on the same surface. Ignore individual readings entirely and watch the seven-day average.
Anchor it to something better. Once a month, run the Navy calculator and note the gap between the two. If your scale runs 5 points below the tape method consistently, that offset is your calibration, and it tells you how to interpret the scale's output going forward. What you must not do is switch between quoting the scale and quoting the tape depending on which is more flattering that week.
Verdict. Good as a high-frequency trend smoother, poor as your only source of truth. If you own one, keep using it; if you do not, buy the tape instead.
Branch D: You Have Gym or Clinic Access
If you already pay for a gym, there is often a better instrument sitting in the corner that you have walked past a hundred times.
Multi-frequency bioimpedance stations. InBody and similar units use hand and foot electrodes and multiple current frequencies, which lets them separate intracellular from extracellular water and segment the body into limbs and trunk. Accuracy runs ±3–5 points, better than a foot-only scale but still hydration-sensitive. Many gyms include a monthly scan with membership. Arrive in a standardized state: no training beforehand, no large meal within three hours, no heavy fluid load.
Skinfold calipers with a trained tester. In competent hands, a three- or seven-site Jackson-Pollock test reaches ±3.5 points. In untrained hands it is worse than a tape measure, because two testers pinching the same abdomen routinely differ by 3–5 millimeters. If the person testing you does it daily as part of their job, this is a good option. If it is a personal trainer who read about it once, it is not. Insist on the same tester every time, since inter-tester variation is much larger than intra-tester variation.
What to do with the result. Use the gym reading as a periodic anchor, roughly every eight to twelve weeks, and keep tracking at home in between. Note the difference between the gym number and your home number and carry that offset forward, exactly as in Branch C.
Branch E: You Have Budget for a Scan
One scan, used correctly, improves every home measurement you take afterward. This is the highest-value spend in the entire list, and it is a one-time cost, not a subscription.
DEXA, $45–150. A low-dose X-ray scan producing fat, lean, and bone mass by region in about ten minutes. Accuracy ±1–2 percentage points, and it is the only common option that reports regional distribution, including a visceral adipose tissue estimate. Book morning appointments, fasted and normally hydrated, and use the same facility for any repeat.
BodPod, $50–75. Air displacement plethysmography. Measures body volume, derives density, converts through the Siri equation. Accuracy ±2–3 points. Faster than DEXA, no radiation, and sensitive to clothing and hair, so follow their prep instructions precisely.
Hydrostatic weighing, $40–60. Underwater weighing, the historical gold standard, ±2–3 points. Requires full exhalation while submerged, which some people find genuinely unpleasant.
The calibration play. Get the scan, then within 48 hours run both the Navy calculator and the photo AI estimator under your standard conditions. Write down all three numbers. If DEXA says 18.5% and your Navy result says 21.5%, your personal Navy offset is +3 points, and every future Navy reading can be mentally adjusted. You have effectively bought reference-grade accuracy for a free method — and you never need to buy it again unless your build changes substantially.
Do not repeat scans often. Twice a year at most, and only if something has genuinely changed. The scan is for calibration, not for tracking.
Putting the Branches Together
Whichever branch you started in, the same 90-day structure applies, and it takes under ten minutes every other week.
Week 0. Establish a baseline with your primary method plus one secondary, on the same morning. Record the raw inputs — every circumference, your weight, the photo conditions — not just the output.
Every two weeks. Repeat the primary method under identical conditions. Duplicate every tape reading and average. Add a context note if you traveled, were ill, slept badly, or are in a phase of your cycle where water retention is higher.
Monthly. Run the secondary method as a cross-check. Compare the direction of both, not their absolute values.
Day 90. Look at the full series. A moderate deficit produces roughly 0.5 to 1.0 percentage points of fat loss per month, so expect 1.5 to 3 points over the quarter. Individual readings will bounce by a point or more in both directions along the way; that scatter is normal and should not trigger program changes.
What counts as a real plateau. Four or more consecutive check-ins with no downward movement, under consistent conditions and consistent habits. Anything shorter is noise wearing a costume.
The branch you pick matters much less than the consistency you bring to it. A photo estimate taken faithfully every two weeks for six months will tell you more about your body than three DEXA scans taken at random intervals when you happened to feel motivated.
Frequently Asked Questions
What is the cheapest way to figure out my body fat percentage?
Free, with a phone: upload front and side photos to the photo AI estimator. For about six dollars, a cloth tape measure and the Navy calculator gives you a second independent estimate with a slightly tighter error band.
Are smart scales accurate enough to rely on?
Not as a standalone number. Bioimpedance scales typically run ±5–8 percentage points and shift with hydration, so use the seven-day average as a trend smoother and anchor it monthly to a tape-based reading.
Should I pay for a DEXA scan?
Once, if it fits your budget. Its real value is calibration: it tells you how far off your free home method runs for your specific body, and that offset makes every future home reading more meaningful. Repeat scans are rarely worth the cost.
Which single method should I use if I only want one?
The Navy circumference method, if you will actually do the taping. Photo estimation, if tape measurements are the kind of chore you will quietly stop doing after three weeks. Adherence beats precision over a six-month horizon.
How often should I check my body fat percentage?
Every one to two weeks for your primary method. Daily checks add noise without adding signal, because real composition change over 24 hours is far smaller than any home method can resolve.
Pick Your Branch and Take the First Reading
You do not need the best method. You need the best method you will still be using in March, measured the same way every time.
Body Fat Estimator covers the two branches most people land in: the photo AI estimator for the phone-only route and the Navy body fat calculator for the tape route, with the visceral fat calculator when abdominal distribution is the real question.
See pricing for unlimited check-ins and saved history. Take your baseline tomorrow morning, write down every input, and put the next check-in on your calendar before you close the tab.
