Ring size tools: making an on-screen measurement mean something

Most on-screen ring sizers are a gesture rather than a measurement. This case study covers two tools built for a fine jeweller, what was wrong with the sizing content they replaced, and the calibration step that’s the difference between a number you can order from and a number that changes with the phone you’re holding.

At a glance

Client A New York fine jeweller
Tools A calibrated ring gauge wrapped in an education article, and a full six-method ring size finder
Scale US sizes 3 to 13 in quarter steps, 41 conversion rows
Methods Six, each with its own stated accuracy
Status Prototypes, pending client decision

The starting position

Three problems, found by reading the live site rather than assuming.

The only ring size chart was a PDF. The sizing page covered how to measure your finger, which sent you off to order a free posted gauge, and how to measure a ring you already own, which was one sentence and a link to a PDF on a CDN. A PDF can’t rank, can’t answer a question in a search result, and is awkward to read on the phone the customer is holding.

Both routes needed something the customer didn’t have yet. The free posted gauge is genuinely the best answer, and it arrives in a few days. The chart needs a printer. There was nothing at all for the person who wants a number in the next two minutes, which is most people, most of the time.

The reassurance was on the wrong page. Complimentary resizing within thirty days is the fact that makes a measurement low-risk. It appeared once, in the middle of a paragraph about how to use the gauge, several hundred words from the answer it should have been sitting next to.

The observation the build hangs on

A package addressed to the house is the single thing somebody planning a proposal does not want, particularly if they live with the person they’re proposing to.

So the jeweller’s most generous sizing service is unusable by a meaningful slice of the audience buying their most important product, and nothing on the site acknowledged that or offered an alternative. That’s the gap the tool fills, and it isn’t a technical gap at all.

Calibration

The only on-screen ring sizer I found published by another jeweller asks you to place a ring on the screen and adjust a circle, with no calibration step anywhere. Without one, the same ring returns a different size on every device, because every screen has a different pixel density.

A bank card is 85.60mm wide under ISO/IEC 7810 regardless of who issued it. Match an on-screen outline to the card, and the browser knows how many pixels make a millimetre on that specific screen. After that, matching a circle to the inside of a ring is a measurement.

Six methods, six honest margins

Method Stated accuracy
Screen gauge against a ring A quarter size
Ruler across a ring A quarter to a half size
Tape around the finger Half a size
Traced outline Half a size
Paper strip A full size
Common sizes Two sizes

Publishing the margin alongside each method matters more than it looks. A customer who knows the paper strip method is rough will treat the answer differently from one who assumes it’s exact, and the tool says plainly that it’s the least dependable method in common use.

The quarter-size scale is what makes those margins usable. The jeweller sells quarter sizes and explains them by worked example on their own site, so an adjustment can be applied honestly rather than rounded to the nearest half and thrown away.

Confidential mode

Same feature, different register from the version built previously for a high street jeweller. The bar reads confidential mode, the cover page is Shipping and Returns rather than anything about sizing, and the tone is discreet rather than conspiratorial. No winking.

Under every result sits a reassurance panel: sizing is complimentary within thirty days, so a quarter size either way is a conversation rather than a problem. That single panel does more to remove hesitation than the accuracy rating above it.

One exception is promoted throughout, because it’s the most expensive thing a customer working from a secret measurement can get wrong. A full eternity band has stones the whole way around the shank, no plain metal to work with, and can’t be sized, so it has to be remade. It appears in the result, in the confidential section and in the FAQ.

What I learned

The best insight on this project came from the returns policy, not the technology. Free resizing within thirty days changes what an imperfect measurement means, and moving that one sentence next to the answer was worth more than any accuracy improvement.

The second learning is that honesty about margin makes a tool more persuasive rather than less. Stating that one method is only good to a full size makes every other number on the page more believable.

Limitations

These are prototypes pending a client decision, so there’s no usage or conversion data yet. The typefaces are substitutes, since the brand faces are licensed and not on Google Fonts, which is stated in the footer so nobody mistakes the prototype for the real thing. A calibrated screen measurement is still a screen measurement, and a posted gauge remains more accurate where the customer can use one.

Services used

Interactive tools and calculators and app and tool development. The jeweller’s wider search work is covered in the Kwiat case study, and the jewellery and luxury retail page covers the sector.

FAQ

Are on-screen ring sizers accurate?

Only if they calibrate first. Without a calibration step against something of a known size, the same ring returns a different size on every device, because screens vary in pixel density. Calibrating against a bank card, which is a fixed 85.60mm wide, makes the measurement real.

Why calibrate against a bank card?

Because every bank card is the same size. ISO/IEC 7810 sets the width at 85.60mm regardless of the issuer, so almost every customer has an accurate reference object in their pocket. Matching an on-screen outline to it tells the browser how many pixels make a millimetre.

What if someone is buying a surprise?

That’s what confidential mode is for. A free sizing kit posted to the house is the one thing a person planning a proposal doesn’t want, so the tool offers a discreet route with a neutral cover page and reassurance about resizing, rather than pretending the problem doesn’t exist.

Can an eternity ring be resized?

A full eternity band usually can’t, because stones run the whole way around the shank and there’s no plain metal to work with, so it has to be remade instead. That makes it the most expensive thing to get wrong when working from a secretly taken measurement.

Got a sizing problem on your store?

Get in touch and I’ll look at what your site currently asks customers to do before they can order.

Get started

Proof, not promises.

Tell me what's going on and I'll come back with what I'd do first. No obligation, and no report you need a translator for.

Over 925,000 impressions and 13,795 clicks in a year, from people asking one awkward question: what size do I need?condoms.uk, past 12 months

Brands and sites I have worked on

Brands worked on across nearly twenty years, in agency roles and directly.