UPC-E barcode generator
The compressed form of UPC-A for packaging that has no room for twelve digits. Eight printed digits that a scanner expands back to a full UPC-A.
Generate a UPC-E barcode
SVG shape mask (decorative)
Upload an SVG — the barcode is clipped to its shape. A horse SVG makes a horse-shaped barcode.
UPC-E is not a separate product number. It is a UPC-A with runs of zeros squeezed out, printed in about half the width. When a till scans it, the reader expands it back into the twelve-digit UPC-A before the point-of-sale system ever sees it, so the product record is the same one.
The compression only works on numbers with the right shape. Which zeros can be dropped depends on the last data digit: values ending 0, 1 or 2 move that digit into the manufacturer field; a trailing 3, 4 or 5–9 each has its own expansion rule. If your UPC-A does not contain a long enough run of zeros in the right place, it simply has no UPC-E form and you must print the full symbol.
Only number systems 0 and 1 are eligible. A UPC-A that begins with 2, 3, 4, 5 or 9 cannot be shown as a UPC-E at all.
What UPC-E encodes
| Character set | Digits 0–9 only |
|---|---|
| Length | 7 digits (check digit computed) or a complete 8 |
| Check digit | Modulo 10, computed over the expanded 12-digit UPC-A, not over the printed digits |
| Dimensions | Linear (1D) |
| Nominal size | 22.11 × 25.91 mm at 100 % magnification |
| Standard | ISO/IEC 15420, GS1 General Specifications |
Where UPC-E is actually used
Small North American retail packs
Confectionery, cigarettes, cosmetics and anything sold in a pack too narrow for UPC-A.
Blister cards
Small hardware and pharmacy items where the printable area is a few centimetres wide.
Beverage cans and bottles
Curved surfaces where a narrower symbol has a better chance of a clean scan.
Rules this UPC-E generator enforces
- Digits only, 7 or 8 of them. The first digit is the number system and must be 0 or 1.
- The check digit is derived from the expanded UPC-A, which is why it cannot be worked out from the eight printed digits by hand.
- Not every UPC-A has a UPC-E equivalent. If the zero pattern does not match one of the compression rules, there is no valid short form.
- Never invent a UPC-E directly. Start from the UPC-A that GS1 allocated to you and compress that.
Printing a UPC-E symbol that scans
UPC-E has the same 80 %–200 % magnification range as UPC-A, from a narrower 22.11 mm nominal width. Because the symbol is already small, the left and right quiet zones — nine and seven modules respectively — matter more here than anywhere else in the EAN/UPC family.
On a curved pack, run the bars parallel to the axis of the curve wherever possible. A symbol wrapped tightly around a small cylinder distorts the bar widths seen by the scanner and reads intermittently even when printed perfectly.
UPC-E questions
Is UPC-E a different product code from UPC-A?
No. It is the same GTIN-12 printed in a compressed form. The scanner expands it before the code reaches the till system.
Why will my number not convert to UPC-E?
Compression requires a run of zeros in a specific position and a number system of 0 or 1. Numbers without that pattern have no UPC-E representation.
How many digits are printed under a UPC-E?
Eight: the number system digit, six data digits and the check digit, with the first and last set outside the guard bars.