The History of ZIP+4: How Four Digits Changed American Mail

1943: postal zones, and the two-digit prefix

The problem arrived with the Second World War. Experienced postal clerks were drafted, and the people replacing them did not carry the geography of a city in their heads the way a career sorter did. Knowing that a particular street ran between two neighbourhoods, and which station served it, was learned knowledge, and the Post Office Department had suddenly lost most of it.

The stopgap was to write the knowledge down. In 1943 the Department assigned zone numbers to 124 of the largest cities, so an address could be written Cleveland 12, Ohio. A new clerk did not need to know where the street was. They needed to read a number and put the envelope in the matching slot.

That is the design principle underneath everything that followed: push the routing decision out of the sorter's memory and onto the envelope itself.

1963: the ZIP Code

Zone numbers only covered big cities, and only the last leg of the journey. By the early 1960s mail volume was climbing steeply — more business correspondence, more direct mail, more bills — and the Department needed a scheme that covered the whole country and described the entire route.

The answer went live on 1 July 1963: the Zone Improvement Plan, marketed with an acronym chosen so the Department could talk about mail zipping along. Five digits, read left to right, from broad to narrow:

The three-digit sectional idea had been proposed years earlier by a postal inspector named Robert Moon, who had been arguing for it since the mid-1940s. The final scheme extended his prefix with two more digits.

Adoption was voluntary at first, and voluntary schemes involving extra effort tend to go nowhere. The Department ran an unusually heavy publicity campaign, including a cartoon mascot — Mr. ZIP, a stick figure postman who turned up on stamp margins, posters, and television. Bulk mailers were brought in harder: within a few years, using ZIP Codes became a condition of the discounted rates that made bulk mailing economical in the first place. That is what actually moved the needle.

1983: ZIP+4

Twenty years on, the constraint had moved. Machines could now read addresses optically, but a five-digit ZIP only got an envelope to the correct post office. From there a human being still sorted it into carrier routes and then into walk order — the sequence a letter carrier actually walks a street. That last stretch was the expensive part, and it was still manual.

ZIP+4, introduced in 1983, encoded that final stretch. Four digits after a hyphen, identifying a delivery segment inside the five-digit ZIP. In practice a segment is:

• One side of one block — say, the odd numbers between two cross streets
• A single large building, or one floor of a building
• A group of PO boxes, and sometimes a single box
• A single firm that receives enough mail to justify its own code

This is why there is no single ZIP+4 for a town, or even for a street. Two neighbours facing each other across the same road usually have different last-four digits, because they sit on opposite sides of the carrier's walk. It is also why a ZIP+4 lookup needs a full street address rather than just a city — the four digits are a property of the delivery point, not of the place.

ZIP+4 was, and remains, optional for ordinary mail. Its leverage was commercial: presort discounts were priced so that mailers who supplied accurate ZIP+4 data paid meaningfully less. An industry of address hygiene grew up around that pricing, and it is still the reason most businesses care about the four digits at all.

The eleven-digit delivery point

One more layer sits on top, and most people never see it because it is printed as a barcode rather than as text. Take the nine digits, append two more that identify the individual delivery point — usually derived from the street number — and add a check digit. Eleven digits, one specific mailbox.

That is the delivery point barcode, and it is what lets automated equipment sort a tray of letters directly into the order a carrier will walk them, with no human handling between the plant and the route. When our API returns a delivery_point_barcode field, this is what it is.

Alongside it came CASS — Coding Accuracy Support System — a certification programme for the software that assigns these codes. If a vendor's engine is CASS-certified, USPS has tested it against known answers and will accept its output for postage discounts. It is the reason address validation is a commercial product rather than something everyone builds in-house.

What the layers tell you

Read the nine digits back to front and you get a history of automation. The first three digits describe where a regional sorting plant is, a 1963 concept built for machines that barely existed. The last four describe a carrier's walking route, a 1983 concept built for machines that could read an envelope but not think about it. The eleven-digit barcode describes an individual mailbox, because by 1991 the machines could handle that too.

Each layer was added when handling costs at that stage of the journey became the bottleneck. None of them replaced the layer beneath. That is why the full code is a hyphenated stack of decisions rather than one clean identifier — and why it still works.

Questions

Keep reading