The lack of technology that sorts and stores the mounds of data collected from cookies and ad tags could contribute to the slow adoption of behavioral targeting, according to some advertising insiders. The culprit becomes the terabytes of data from hundreds of thousands of ad impressions collecting geographic location, content on page, time of day, interaction with ads, frequency in which ads serve up, and more. "There needs to be a technology solution out there that fixes that," says Raleigh Harbour, vice president of business development at Rubicon Project, Los Angeles. "One publisher alone collects an immense amount of data that needs to be unlocked to add more value. That's the only way to target campaigns to the correct audience." So, how do you filter the data to use what you need and dump the rest? What data points are more valuable than the others? And how do you determine the data's worth once you identify its value? Some advertising execs believe aggregating the data and providing specific points in the process to filter content will become the key to better behavioral targeting. Filtering will help the online ad industry best leverage the data collected. Standards could help, too. Today there are as many ways to filter data and assign values as there are Right Medias and DoubleClicks, Harbour says. And while execs complain about the lack of one manageable and consistent method, they might look toward other industries that have experienced the same data overload from an emerging technology. Take radio frequency identification, for example. Those little semiconductor tags, known as RFID, that Wal-Mart Stores, for example, affixed to cartons and pallets to track goods moving through their supply chain collected an overwhelming amount of data -- until Wal-Mart and other retailers' IT departments discovered a way to filter out the important data and purge the rest. Deleting unwanted data becomes the first step in designing a better behavioral targeting platform, Harbour says. After collecting the data, the system would need to determine a price. Then the problem then turns from data quantity to quality. This means having a place in the process where data is measured. It means weeding through bits and bytes to find the best information to turn transactions into a positive return on investment (ROI). This will help determine the data most relevant and associate a value to it. "I'm a big believer in [the] market clearing prices," Harbour says. "It's important to look at historical trends to define the pricing." Stay tuned.