EWeek has a substantive article pointing out all the challenges with RFID -- just in case you need to read about them one more time!

But buried in the article I found these two useful nuggets, about the enticing business case for implementing RFID, for retailers and suppliers:

A study by consultants A.T. Kearney Inc. found that a retailer with an RFID system throughout its organization could expect a one-time cash savings of about 5 percent of total inventory, thanks to improved customer-demand Forecasting; an annual reduction in store and warehouse labor expenses of 7.5 percent; and a yearly recurring gain of $700,000 per $1 billion in annual sales thanks to fewer out-of-stock items and less theft. All told, according to Kearney, a retailer with a wall-to-wall RFID system---which includes readers downloading actionable information, literally by the second, into middleware connected to networked corporate databases---could save 32 cents on every sales dollar. Not a bad return, considering that most big retailers' net margins aren't much better than 3 percent.

Eventually, suppliers should achieve significant benefits as well. Thanks to erratic forecasting and a lack of precise information, out-of-stocks of the top 2,000 items sold in retail outlets are estimated to average about 10 percent currently. If real-time RFID data diminishes out-of-stocks by even 50 percent, as is expected, suppliers could realize "a revenue gain of 5 percent that most companies would kill for," says Larry Kellam, former director of supply-chain innovation at Procter & Gamble (where he was an EPC pioneer), and currently the principal at the Kellam Group, an RFID consultancy.

I suspect it will take years for a company to reach these impressive levels of savings. All the more reason to get started on those pilots now.