How our bra size calculator works

Every formula, threshold and rule the calculator uses, why we chose it, where it falls short, and a dated record of every change.

The short version

We take four measurements and convert them to inches.
We check them against each other. Impossible combinations stop the calculation and ask you to re-measure.
We calculate the band two ways, modern and traditional, and show both.
The cup comes from the larger of your two bust measurements, one cup per inch.
Results that fall midway between two cups show both as equally likely.
A confidence score reports how consistent your measurements are.
All of this runs in your browser. Your measurements are never sent to us.

The four measurements

Measurement	How it’s taken	What it’s used for
Snug underbust	Tape firm around the ribcage, directly under the breasts, breathing out	Band size (both methods)
Loose underbust	Same place, tape resting, breathing normally	Tie-break for exactly odd bands; confidence
Standing bust	Upright, tape level around the fullest part	Cup size, if larger than leaning; confidence
Leaning bust	Bent forward to about 90°, tape loose around the fullest part	Cup size, if larger than standing; confidence

We also ask what you wore while measuring (a padded bra lowers confidence), whether one breast is noticeably larger (adds fitting guidance) and, optionally, your breast shape (used only for style suggestions, never for the size).

Values are entered in steps of 0.25 inches or 0.5 cm. Centimetres are converted to inches for every calculation:

inches = centimetres ÷ 2.54 (rounded to 3 decimal places)

Why four measurements

A two-measurement calculator assumes the tape under your bust tells the whole story about your band, and that a standing bust measurement captures all your breast tissue. Neither is always true.

The loose underbust shows how much soft tissue compresses under the tape. The leaning measurement lets tissue fall away from the chest wall so the tape can pass around all of it, which tends to matter more at fuller cup sizes. We use whichever bust measurement is larger.

This approach draws on the fitting practice of taking more than one bust measurement. We know of no published study validating this exact combination against professional fittings, and we list that under known limitations. Published research has found that the traditional method of adding inches to the underbust frequently does not produce a well-fitting bra, and that the measuring method affects the size obtained. That is why we show both band methods and how each was calculated.

Checking the measurements first

Before calculating, each measurement is checked on its own:

Underbust measurements must be between 20 and 70 inches (50.8–177.8 cm).
Bust measurements must be between 22 and 90 inches (55.9–228.6 cm).
If a value is outside the range but would make sense in the other unit, we ask. For example, “28 cm is too small for a snug underbust. Did you mean 28 inches?”

Then they are checked against each other. Any of these stops the calculation, with an explanation, rather than guessing:

All four measurements are identical.
The loose underbust is smaller than the snug underbust.
Either bust measurement is smaller than the snug underbust.
The bust is more than 30 inches larger than the underbust.

Band size

Modern method (recommended)

band = snug underbust rounded to the nearest even number
if snug underbust is exactly odd (within ±0.005 in): if loose underbust − snug underbust ≥ 3 in: round down otherwise: round up

UK bands are made in even sizes, so the snug underbust is rounded to the nearest one. When it lands exactly on an odd number, the loose measurement breaks the tie: a gap of 3 inches or more means soft tissue compresses a lot under the tape, and the firmer band is likely to sit better.

Traditional method

u = snug underbust rounded to the nearest whole inch
band = u + 4 if u is even
band = u + 5 if u is odd

This older method is still used by some brands and size charts. It gives a bigger band and a smaller cup letter than the modern method, so we show it alongside, never instead.

Band range

Bands are limited to 28–56, the range UK retailers generally stock. A calculated band outside that range is shown at the nearest end, the cup is recalculated against the band actually shown, and the result carries a note explaining what happened and where to look for other sizes.

Cup size

bust figure = larger of (standing bust, leaning bust)
cup difference = bust figure − band (inches, 2 decimal places, per method)
cup = UK cup ladder[ round(cup difference) ]

Cup difference (in)	Cup difference (cm)	UK cup
0	0	AA
1	2.5	A
2	5.1	B
3	7.6	C
4	10.2	D
5	12.7	DD
6	15.2	E
7	17.8	F
8	20.3	FF
9	22.9	G
10	25.4	GG
11	27.9	H
12	30.5	HH
13	33	J
14	35.6	JJ
15	38.1	K
16	40.6	KK
17	43.2	L
18	45.7	LL

Between sizes

If the fractional part of the cup difference is between 0.4 and 0.6 inches inclusive, we return both neighbouring cups as equally likely instead of rounding. A difference of 4.5 inches on a 32 band gives 32D and 32DD. Rounding would hide what is really a coin toss.

Edge cases

Negative cup difference: shown as AA, with a note suggesting soft-cup and bralette styles.
Beyond K: flagged, with a pointer to full-bust specialist brands.
Beyond LL: shown as LL+, with a recommendation for a professional fitting.
Small band with a larger cup: a band of 30 or smaller with a cup of F or above carries a note that high-street availability is limited.
Padded bra: a warning that the result may be about one cup too large.
Uneven breasts: guidance to fit the larger side.

Confidence scoring

Each warning sign adds a fixed weight. Nothing ever raises the score.

Signal	Threshold	Weight
Snug and loose underbust far apart	loose − snug > 4 in	1
Leaning bust much larger than standing	leaning − standing > 5 in	2
Standing bust larger than leaning	standing − leaning > 1 in	2
Measured over a padded bra	as reported	2
Between two cup sizes	fractional difference 0.4–0.6 in	1
Band outside the UK range	below 28 or above 56	1
Cup beyond K	modern-method cup	1

total 0 → high confidence
total 1–2 → moderate confidence
total 3+ → low confidence
any error → no size, re-measure

Every signal comes with a plain-English reason shown in your results. Confidence checks your measurements against each other, not against your body: if the tape is misplaced the same way every time, all four can agree and still be wrong.

Sister sizes

sister size = (band − 2, cup + 1) tighter band
= (band + 2, cup − 1) looser band

We show up to two sister sizes in each direction from the modern-method size, within 28–56 and AA–LL, each labelled with the fit problem it helps with.

International conversion

Conversions come from a table stored with the site. Bands 28–44 and cups A–K come from our launch table. Other rows follow the same arithmetic pattern and are marked wherever they appear. Cups beyond K have no standard international equivalent and are not converted. The table has not yet been checked against current retailer size charts, and the converter says so.

Cup volume comparison

volume step = band ÷ 2 + cup position (AA = 0, A = 1, B = 2 …)

Sizes on the same step are sister sizes; each step up is roughly one cup size more volume. This is an approximation within the sizing system, not a measurement of breast volume.

Brand translation

Each brand record stores its size range, whether its chart follows the modern or traditional method, any band adjustment (in inches, where +2 is one band up) and cup adjustment (in cup steps), and the date we checked it against the brand’s published size chart. A brand without a verification date never produces a size: the site shows “Sizing data pending verification” instead. 0 of 12 brands are currently verified.

No-tape estimates

From a bra you own

Band too loose, on any hook: one band down, cup letter up one to keep volume.
Band tight on the loosest hook: one band up, cup letter down one.
Band tight on a tighter hook: same band, with one band up as an alternative.
Band right only on the tightest hook: same band, with one band down as an alternative.
Band rides up: one band down, unless the band also feels tight, which we flag as a likely cup problem.
Cups overflowing: one cup up, with two up as an alternative.
Cups gaping: one cup down, with the current cup as an alternative, since gaping can be a shape issue.
Straps dig in: one band down as an alternative. Straps slip: a style note only.

From clothing size

Editorial heuristic, not retailer data. UK dress sizes do not map reliably onto bra bands: clothing is cut for the bust, waist and hips, not the ribcage. Each dress size lists a deliberately wide spread of bands so the estimate is a starting point for trying bras on, not a size. Documented on /methodology.

UK dress size	Bands suggested
6	28, 30
8	28, 30, 32
10	30, 32, 34
12	32, 34, 36
14	34, 36, 38
16	36, 38, 40
18	38, 40, 42
20	40, 42, 44
22	42, 44, 46
24	44, 46, 48
26	46, 48, 50
28	48, 50, 52
30	50, 52, 54
32	52, 54, 56

Your current cup letter is assumed to go with the middle band. The other bands use sister sizes so the cup volume stays the same.

What the AI tools do, and don’t do

Three tools use AI: fit diagnosis, style suggestions in your results, and questions during step-by-step measuring. The AI never calculates your size. Sizes always come from the formulas above, running in your browser.

Models: claude-haiku-4-5-20251001 for single fit problems, style suggestions and measuring questions; claude-sonnet-5 for combinations of fit problems.
Responses are limited to 1000 tokens and must match a fixed structure.
Our code then checks every answer: sizes must be real UK sizes, brands can only come from verified data, and safety notes from our fixed rules are always kept.
The AI receives only fit problems, sizes, breast shape, style need and typed text with numbers removed. Never measurements.
If the AI is unavailable or a visitor has made more than 20 requests in an hour, the tool serves our standard guidance, built from the same fitting rules.

Known limitations

No validation study for this exact method. We know of no published study testing this combination of four measurements against professional fittings.
Technique matters most. A tilted or twisted tape, or a snug measurement that isn’t firm, changes the result. One centimetre is about 0.4 of a cup.
The cup ladder is a convention. Brands don’t all follow it, which is why brand translation exists and why unverified brands show no conversion.
Bra size isn’t breast volume. Research has found bra size to be a poor indicator of breast volume, so the volume comparison is an approximation.
Shape and style. The same size can fit differently in different styles and on different shapes.
Uneven breasts. One size has to serve two breasts; we recommend fitting the larger side.
Estimates are heuristics. The no-tape tools use published rules of thumb, not measurements.
Bodies change. A size is right for now. Pregnancy, weight change, menopause and HRT can all change it.
AI suggestions can be wrong. They are checked against fixed rules, but not guaranteed.
Not medical advice. Pain, numbness, skin damage or breast changes need a GP, not a new size.

Sources

White J, Scurr J (2012). Evaluation of professional bra fitting criteria for bra selection and fitting in the UK. Ergonomics.
Greenbaum AR, Heslop T, Morris J, Dunn KW (2003). An investigation of the suitability of bra fit in women referred for reduction mammaplasty. British Journal of Plastic Surgery.
Pechter EA (1998). A new method for determining bra size and predicting postaugmentation breast size. Plastic and Reconstructive Surgery.
McGhee DE, Steele JR (2010). Optimising breast support in female patients through correct bra fit. A cross-sectional study. Journal of Science and Medicine in Sport.
McGhee DE, Steele JR (2011). Breast volume and bra size. International Journal of Clothing Science and Technology.

Changelog

Version 1.0.0: Initial published methodology

15 September 2026

Four measurements: snug underbust, loose underbust, standing bust and leaning bust.
Modern band (snug underbust rounded to the nearest even inch, with a soft-tissue tie-break) and traditional band (+4 / +5) calculated side by side.
Cup calculated from the larger of the standing and leaning bust measurements against the UK cup ladder, AA to LL.
Results whose cup difference has a fractional part between 0.4 and 0.6 inches return both adjacent cups as equal recommendations.
Confidence scoring based on the internal consistency of the four measurements.

Use the tools

Sister Size Calculator

Every size with the same cup volume, and when to try each.

Bra size calculator

Four measurements, both band methods, sister sizes and a confidence score..

Fit Diagnosis

Select your fit problems and get a specific size to try.

Frequently asked questions

Leaning forward lets breast tissue fall away from the chest wall, so the tape can pass around all of it. A standing tape can miss tissue that sits flat against the chest. Using the larger measurement avoids undersizing the cup; if the two differ a lot, the confidence score drops and says why.

UK bra bands are made in even sizes, so the calculator rounds to the nearest even number rather than showing a band nobody sells.

Not this exact combination, as far as we know. It draws on the fitting practice of taking more than one bust measurement, and published research shows the older +4 method often gives a poorly fitting bra. We list the lack of a validation study under known limitations.

Each warning sign adds a fixed weight: minor signals add 1 and major ones add 2. A total of 0 is high confidence, 1 or 2 is moderate, and 3 or more is low. The thresholds and weights are listed on this page.

Every calculated result carries the version it was calculated with. It appears on printed size cards, downloaded images, share links and results saved to your device, so you can always tell which method produced a size.