The Road to
Industry in the 1960s
Computerization of the U.S. grocery industry (consumer goods manufacturers, wholesalers and retailers) starting in the early 1960’s, involved dealing with an already efficient distribution channel. However, it was generally agreed that additional opportunities for improvement existed. Among these were three that became the basis for the development and implementation of the UPC system:
Reducing the labor cost of retail checkout. Roughly 40% of all retail store labor expenditures went to checkers and baggers.
Capture of individual item movement data at point of sale. Given the limitations of mechanical cash registers, the only item information which could be captured was the price paid and which of 5-6 departments (e.g. meat, produce) the item was found. Individual item movement was tracked from warehouse to store (i.e. warehouse withdrawals) but with no practical way of adjusting for variations in retail inventory levels.
Allow computer to computer communication of supply chain transactions (orders, shipping advices, invoices, etc.) between retailers and manufacturers. Each participant had their own product code but manually translating back and forth generated a significant error rate which impacted inventory levels and slowed transaction speeds.
Defining a product identification was important in two strong focuses of that time:
creating an identification scheme that improved the retailers ordering of products from manufacturers and the accounting of it; and
automation of the checkout process.
The first effort had only a modest payback. The second effort has saved consumers billions of dollars with a more productive distribution channel.
During the 1960s independent projects began to try to improve productivity at checkout by changing the key entry of price to some optical automation. Early on, Sylvania worked with Stop & Shop in Boston attempting to read the packaging information on items at checkout using an incandescent light scanner, but the amount of light required and the lack of structure in the information on packages made this impractical. In 1966 in Cincinnati, OH the Kroger Company realizing that military contractors were developing sophisticated technology, sponsored a multiple day conference of high tech manufacturers to educate them on the grocery distribution environment and discuss opportunities to apply that technology to grocery distribution. RCA attended that conference and approached Kroger about assuming Kroger's project with Sylvania and developing a laser scanner to operate at the checkout. RCA had purchased the bulls-eye bar code patent sold by Woodland and Silver to Philco years earlier. By 1968 RCA had a prototype laser scanning checkout running in their Princeton, NJ laboratory and used it to measure performance and test different checkout configurations.
It's possible that in that same year an executive at Procter & Gamble, also located in Cincinnati, learned about the cross-town Kroger-RCA project. Manufacturers likely would be asked to apply any symbol required and they wouldn't want to have to apply different symbols for different checkout equipment or retailers. The P&G executive likely encouraged the trade associations to work to find a common solution for the ordering/accounting effort and the checkout effort. By the end of the 1960s there had been a lot of talk and quite a few meetings between individual retailers and grocery manufacturers and by the Administrative Systems Committees in both the NAFC (National Association of Food Chains) and the GMA (Grocery Manufacturers Association) about ways to improve the product ordering process. Little progress was made. Most accounting used the existing case code created by the manufacturer. Manufacturers favored adding a prefix when needed to get an item code. In the late 1960s various industry trade groups, Clancy Adamy of the NAFC, Mike O’Connor of the SMI (Super Market Institute now called the FMI or Food Marketing Institute) and the GMA held several conferences and meetings each ending with little consensus. Retailers still wanted seven or less characters in the code and manufacturers felt they needed eleven or more. Clancy Adamy asked Andy Pearson at McKinsey and Associates to suggest a solution. While preparing for a response, McKinsey uncovered the Kroger/RCA project and recognized defining a retail product ID had much more complex requirements. Unfortunately NAFC was not prepared to fund such a study at that time. What did come out of all these meetings was an agreement that a separate stand alone group should be formed to reach the solution. This was the start of the Ad Hoc Council.
Back to the start.
More about the Ad Hoc in the next installment.