Blog
August 29, 2026
-
7 min read

Six rectangles

DBC1 Team

Product & platform

Focuses on measurement, attribution, and turning card usage into something ops and revenue teams can trust.

Put a business card from Tokyo, Warsaw, Chicago, Sydney, Shanghai and Munich on a table and square them up at one corner. They almost line up. Almost.

91×55. 90×50. 88.9×50.8. 90×55. 90×54. 85×55.

Six rectangles, spread across about six millimetres. Not different enough to be a design choice, not similar enough to be the same object. Which is a strange thing to find in a product that has been mass-produced globally for over a century.

The obvious question — why didn't the world just pick one? — turns out to be less interesting than the answers to why each one. Because the six sizes come from six unrelated places: a French photography studio, a Japanese carpentry measurement, the width of an inch, and in at least one case, nowhere anyone can point to.

Before the sizes, there were two different cards

The modern business card is a merger of two ancestors that had nothing to do with each other.

The first was social. Calling cards, or cartes de visite, moved through European aristocratic life from the seventeenth century onward: you sent one ahead, a servant carried it to a servant, and an elaborate etiquette grew up around when to leave one, whether to fold a corner, and what folding it meant. This card said nothing about what you did. It said who you were, and that you knew the rules.

The second was commercial, and considerably more useful. Trade cards became popular at the end of the seventeenth century in Paris, Lyon and London, and they carried something modern cards don't: maps. Street addresses as we know them didn't exist yet — no house numbering — so a merchant's card had to explain how to physically find the shop. Directions like over against Middle Row, Holborn did the work a postcode does now. The trade card was a business card and a wayfinding device in one, and it was sized to fit in a coat pocket or a purse.

Then house numbering arrived, the map came off the card, and the two traditions collapsed into one object: something that told you who a person was and how to reach them.

A note on the deeper history: nearly every article online traces the business card to fifteenth-century China. It may well be right — but the sourcing is a closed loop of printing-company blogs citing each other, and at least one of them calls the Chinese cards meishi, which is a Japanese word. I'd treat the fifteenth-century date as a plausible story rather than an established fact.

Japan: 91×55, because of carpentry

The Japanese standard is yon-gō — size 4 — and it's the most precisely explicable of the six.

The account given by the Japanese printing trade goes like this. In 1854 the French photographer Disdéri devised a card-mounted portrait format, roughly 82×57mm, and it became the reference for card sizes across Europe. Japan adopted the idea in the nineteenth century but not the units: measurement ran on shaku and sun, not millimetres. The nearest clean value to 82mm was 3 sun — about 91mm. The short edge was set at 1 sun 8 bu, about 55mm. When Japan converted to metric, the ratio was simply frozen and rounded into 91×55.

There's a competing theory that it derives from the shiroku-ban book format, 188×128mm, quartered — which lands in almost exactly the same place. Both routes arrive at a card whose dimensions are, underneath the millimetres, a traditional Japanese carpentry measurement.

Two footnotes make this size my favourite of the six.

The first is that Japanese printers commonly explain the proportions by invoking the golden ratio. 91:55 gives 1:1.654. The golden ratio is 1:1.618. It's close, and it's clearly a justification found afterwards rather than a rule followed at the time — which is a very human thing to do with a number you've inherited and can't change.

The second is that the same card has two names. It's 4-gō in Tokyo and 9-gō in Osaka, identical dimensions, because two regional paper manufacturers gave the same stock different product numbers and neither backed down. That naming split is still in commercial use today.

And one piece of social history that isn't a footnote at all. Japanese cards used to be sized by gender: men used 4-gō at 91×55, women a smaller 3-gō with rounded corners at 85×49, with further distinct sizes used by women in the karyūkai. This persisted until the Equal Employment Opportunity Act came into force in 1986, after which the single 91×55 size became standard for everyone. The card size was a status marker, and it stopped being one by law, within living memory.

Europe: 85×55, and the credit-card story is nearly true

Ask why European cards are 85×55 and you'll be told it's the credit card size, so it fits a wallet.

Check the numbers and it doesn't hold. The ISO/IEC 7810 ID-1 format — bank cards, ID cards, everything in the slots of your wallet — is 85.60 × 53.98mm. The European business card is 85×55: about half a millimetre narrower and a full millimetre taller. Close enough to share a card holder, not close enough to be the same standard.

The chronology is the bigger problem. The European format long predates payment cards being something everyone carried. So wallet compatibility isn't the origin of 85×55. It's a happy coincidence that later became the explanation, because the explanation is satisfying and nobody checks.

Worth noting: there is a genuine ISO paper size in this territory. A8, at 74×52mm, sits in the standard A series. Essentially nobody uses it for business cards. A rare case of a formal standard losing outright to a customary one.

The United States: 3.5×2, and nobody can tell you why

Here's the honest position: there is no regulated standard business card size in the US. 3.5×2 inches — 88.9×50.8mm — is purely a de facto convention.

The usual explanation is again the credit card. It's worse here than in Europe, because the numbers point the wrong way: a US credit card is 3.375×2.125 inches. The business card is wider and shorter. It isn't a match, it isn't an approximation, and payment cards weren't ubiquitous when the format settled anyway.

What's actually distinctive about the American size is its proportion. At 1.75, it's the most elongated of the group apart from Eastern Europe's 90×50 at 1.8. Every metric-world card sits between 1.54 and 1.67. The US card is visibly a different shape, and if you've ever felt that an American card looks slightly letterboxed next to a German one, that's why.

My best guess is the least romantic one: it's two round numbers in inches, it fit the presses and the pocket, and it stuck. Sometimes the reason a standard exists is that somebody had to pick something.

The rest of the map

90×55 — Australia, New Zealand, and the Nordics. 90×54 — China, Hong Kong, Singapore. 90×50 — a huge and underappreciated bloc: Poland, Czechia, Slovakia, Hungary, Russia, Ukraine, Romania, Bulgaria, the Baltics, and, well outside Europe, Brazil, Mexico and South Africa. India uses both the US size and 90×50, depending on who you ask. Iran runs its own 85×48.

The 90mm width recurs across three of the six. It's the closest thing to a global convergence anyone achieved, and it still couldn't settle on a height.

Why they never merged

Here's the part that actually explains the persistence, and it has nothing to do with culture.

The constraint isn't the card. It's everything built around the card.

A Japanese printer will tell you plainly that submitting artwork at 90×55 instead of 91×55 costs extra, because the guillotines are set up for 91×55 and a one-millimetre deviation means a different cutting process. Multiply that by card holders, wallet slots, Rolodex trays, presentation sleeves, plastic laminating pouches and scanner-app crop boxes, and you have a full ecosystem calibrated to a specific rectangle.

Once the holders exist, the card can't move. The card is the cheap part; the infrastructure around it is the expensive part, and the infrastructure votes.

That's why six formats separated by a few millimetres have survived a century of globalisation, standardisation bodies, and an entire industry that would have preferred one SKU. Not because anybody defended them. Because nothing ever made it worth the cost of changing.

So: a French photographer's portrait format, filtered through a Japanese unit of length that predates the metre. A European size that everyone attributes to credit cards that hadn't been invented. An American size with no documented origin at all. And a bloc of a dozen countries quietly agreeing on 90×50 without anyone writing it down.

Six rectangles, six millimetres apart, each one carrying the fingerprints of how it got there.