Evan Schuman's StorefrontBacktalk

The men who invented barcodes for scanning groceries—and then watched their idea sit for decades before it was put to use—are being inducted into the National Inventors Hall of Fame. Joseph Woodland and Bernard Silver came up with the idea of a row of parallel bars to identify products in 1948, and then immediately changed the format to a bullseye shape, which they thought would work better.

But it took 25 years (most of which Woodland worked for IBM, which wasn’t much interested in the invention) for optical scanning technology to catch up so the idea could be adopted as UPC codes in 1974. And even though barcodes were on their way to becoming the most important technology in retail, by 1976 BusinessWeek pronounced it a failure. Analysts had predicted 1,000 stores would have scanners by then, but only 50 did. Clearly, barcodes were a flop. A decade later they were everywhere.

Barcodes got their start when the president of a Philadelphia grocery chain, Food Fair, was pleading with a dean at Drexel University to do research into automating the grocery checkout process. The dean said no, but graduate student Silver overheard the request. He told Woodland, another grad student, who came up with the idea of parallel bars to encode identifying data. (Curving the bars around into a bullseye was intended to make the barcode readable from any direction.)

Woodland and Silver built a scanner from a 500-watt lightbulb and parts from a movie projector. It worked, but it wasn’t practical. Woodland went to work for IBM in 1951 and offered the company the invention, but Big Blue didn’t want to pay Woodland and Silver’s asking price. Eventually RCA bought it, and in the spring of 1971 the company demonstrated a bullseye barcode scanner at a grocery trade show. An IBM marketing guy saw the demo, remembered that IBM had the barcode’s inventor on staff, and IBM started work on its straight-line barcode version, with Woodland on the development team.

It turned out that the bullseye wasn’t such a good idea; on test labels, the ink tended to smear, making the label unreadable. When a straight-line barcode’s ink smeared, it just made the lines longer. IBM’s barcode became the UPC in 1974, but the laser scanners to read it were still so expensive—and consumers and lawmakers were so resistant—that it took until the 1980s, when Kmart and other big chains started using them, for barcodes to finally make it big, a mere 40 years after they were invented.

Keep that in mind the next time you hear from analysts that a hot new retail technology is about to take over the world—or that it’s a flop. It may just take a little longer than anyone expects.