Guide
Four methods, ranked honestly by how accurate they are, followed by the six factors that move your size after you’ve measured it.
A ring has a hard, circular inner edge — the one thing in this whole process that is easy to measure precisely. Place it on a calibrated screen and match the circle to the inside edge, or measure across the opening with digital callipers.
Make sure it fits the finger you’re actually buying for. Ring sizes differ between hands and between fingers on the same hand.
Cut a strip of paper about 5 mm wide, wrap it snugly around the base of the finger, mark where it overlaps, and measure the marked length against a calibrated ruler. That length is the inside circumference.
Paper doesn’t stretch, which is exactly why it beats string or thread. Pull it snug, not tight — you are measuring the finger, not compressing it.
A printed strip with sized holes and a wrap-around gauge. The failure mode is entirely in the printer: “fit to page” quietly shrinks the sheet by a few percent and every size on it becomes wrong.
Always verify the printed scale-check square with a ruler before trusting the sizes.
Useful when you have nothing else. Lay the finger flat between two calibrated guides and read the width.
A finger is oval, not round, and soft tissue compresses. Treat the result as a starting point and confirm it another way.
After you measure
A measurement is a number about your finger. A ring size is a decision about a specific ring on that finger. These are the gaps between the two — and our sizer applies every one of them for you.
A wide band contacts more of your finger and wears tighter than a thin one at the same internal diameter. Below about 4 mm it makes no practical difference; at 6 mm expect to go up about a quarter size, and at 8–9 mm about half.
Comfort-fit bands are domed on the inside, so only a narrow ridge touches your finger. They slide over the knuckle more easily and run large — you normally size down.
The ring has to clear the knuckle, then sit at the base without spinning. When the knuckle is clearly the wider of the two, size for the knuckle and use sizing beads or a ring guard to stop the spin.
Fingers are smallest cold and first thing in the morning, and largest when warm, after exercise, or at the end of the day. Measure in the evening at normal room temperature and you’ll land on a size that fits for most of your waking hours.
A salty meal, a hot day, a long flight, pregnancy and some medications all cause temporary swelling. Don’t size a permanent ring on a day when your hands feel puffy.
The dominant hand is usually slightly larger. Measure the exact finger on the exact hand the ring will live on.
Accuracy
Every measurement has an error, and a tool that hides it is not being straight with you. Ours comes from two independent sources, which we combine in quadrature — the standard way of adding uncertainties that are unrelated to each other.
Calibration error is proportional: if your screen scale is 0.5% out, every measurement made on it is 0.5% out. A card gives about ±0.5%, a coin about ±0.8% because the same pixel error is a larger fraction of a smaller object, and two references that agree with each other get you to about ±0.3%.
Measurement error is absolute: matching a hard ring edge is good to roughly ±0.15 mm, a paper strip to about ±0.22 mm of diameter, and a bare finger to around ±0.5 mm because there is no hard edge to align to.
One US size is 0.8128 mm of diameter, so a total error of 0.16 mm is about a fifth of a size. That is roughly the best a screen sizer can physically do — and it is why our top confidence tier says “high”, not “exact”.
The practical takeaway: measure an existing ring rather than a finger, calibrate with two different objects, and don’t zoom the page afterwards. Those three things move you from ±0.65 of a size to about ±0.4.
Both stretch. You will read a size too large, then wonder why the ring spins.
A tilted card is optically shorter, so you calibrate small and every result comes out small.
Browser zoom changes the size of a CSS pixel and silently invalidates your scale. Our sizer watches for this and warns you.
If you measured a 2 mm band and you’re buying an 8 mm one, you need about half a size more.
Come in from the cold, wait ten minutes, then measure.
“Fit to page” is the single most common cause of a wrong printed size.