The warning comes as a number of manufacturers are delaying RFID roll-outs, or only doing the minimum amount of work possible, because they cannot see enough benefit to justify the investment.
"To successfully pilot and implement the change, evidence has shown that manufacturers implementing RFID must start the process now," said Tom Ryan, vice-president of value chain research at Aberdeen Group and the author of the report.
"Many are adopting a hold-back strategy, letting others solve the inevitable problems and pass on the knowledge they will acquire. However, with the tight timelines for compliance, this strategy is exposing manufacturers to real compliance failure."
Manufacturers such as Nestl' have raised fears that retailers stand to benefit most from RFID, and have questioned the return on investment they could receive from meeting the mandates set by companies such as Tesco, Wal-Mart and Metro Group.
However, Ryan said manufacturers that embraced RFID now would reap the most benefit, although he warned that they must be willing to work through the challenges of a new technology, anticipate set-backs, and manage road-blocks.
Despite its benefits for tracking goods, RFID is not proving particularly popular in the delivery market, said Graham Nugent, strategic IS manager at UPS.
"We are still not using RFID in our package business, although we are doing specific projects with customers on the supply chain side," he said.
"The issue for us is what benefits we will get, as we get everything we need from the smart barcodes we have developed," Nugent said.
MIT sets up group
The Massachusetts Institute of Technology's Auto-ID Lab has formed the Web Services Wan Special Interest Group, which will devise a way of sharing RFID data over wide area networks using web services standards.
"No one had pinned down how you share data over a Wan," said Stephen Miles, the organiser of the group. "It is not an issue until you start looking at sharing real-time data with multiple partners."
The challenge is to create a Wan to share data among thousands of simultaneous users.