Analysis: Approaching RFID with your business head on

Radio frequency identification has become a hot concept, promising to streamline how businesses track and stock inventory.

But companies may need to rethink their software infrastructures in order to make RFID work as advertised, say analysts and technology makers.

RFID tags, which combine chips that carry descriptive information and radio frequency technology to track inventory, could make it easier for businesses to track products and raw materials, and therefore reduce their operating costs over time. And with heavy hitters in the retail and consumer packaged goods industries, such as Wal-Mart and Gillette, already pushing partners to develop RFID capabilities, adoption of the tools appears inevitable.

Already the US Department of Defense has said it will require suppliers to use RFID. And on Wednesday, the Food and Drug Administration issued a report urging RFID adoption among drug wholesalers, manufacturers and retailers. The technology is "needed to secure the integrity of the drug supply chain," according to a statement from the agency.

Early resistance to RFID adoption has come from civil liberties groups, which fear that the technology could lead to unprecedented surveillance of consumers. But industry watchers and technology vendors have identified a more mundane potential problem for RFID adopters. They warn that in the rush to launch RFID projects, businesses may be overlooking a crucial element necessary to allow the technology to work smoothly: Making sure back-end databases and business applications can handle the massive amounts of information generated by RFID-enabled systems.

"Companies are going to have problems when they drop RFID on top of shaky infrastructures," said Kara Romanow, an analyst at AMR Research in Boston. "In order to do RFID right, to see a true return, the first thing [a company] needs to do is finish a data synchronisation initiative, and do it right."

Data synchronisation means organising the information stored in disparate company databases - often from different vendors - to make sure those databases are speaking the same language. For instance, one might calculate shipping costs in euros while another uses dollars. "Data synchronisation is the hard work that these companies will have to do for themselves, especially when you talk about changing the data itself and rationalising the data," Romanow said.

Matt Hines writes for CNET News.com.