Enterprises reap benefits from tagging data

Next-generation search technologies for the Internet can also apply to the enterprise for organizing -- and simply finding data -- on their own intranets.

Businesses that realize the importance of tagging and classifying their data will gain an advantage over those that don't, said Clare Hart, an executive vice president at Dow Jones & Co. and chairman of Factiva, the company's subscription business news and information service.

Case in point: Hart said a client recently told her of an employee who was writing a detailed report on a prospective corporate customer. Two weeks into the project, the employee found a 20-page report on the same company on a printer.

"This person was beside themselves," Hart said. "It happens everywhere."

New search technologies, such as tagging, mean less time wasted searching for difficult-to-find items or those that have simply fallen in dark, digital obscurity on company networks.

This week, researchers gathered at the W3C (World Wide Web) meeting in Edinburgh, Scotland, have heralded an approaching era in which computers will be able to interpret descriptions of data, making intelligent links between that data for ever sharper, relevant searches -- a semantic Web.

The idea also applies to company intranets. But to take advantage of the benefits, enterprises should first form a strategy for managing their information, taking an inventory of their own data and figuring out who needs access to it, when and how, Hart said.

The next step is creating taxonomies to classify the data that will be searched. Enterprises are increasingly adding metadata to their data, and many have developed several taxonomies, Hart said.

Taxonomies vary from business to business. Categories could include sales proposals, competitive intelligence reports and market research.

Factiva, the online business, news and information service, uses automated tagging for about 80 percent of the 5 million items it collects and indexes for its subscription service. Information is classified by company, industry, region, subject and language, Hart said. The remaining 20 percent is classified manually.

Factiva uses search engine technology from Fast Search & Transfer ASA and Verity Inc., a company acquired by Autonomy Corp. PLC last November, both of which support taxonomies, Hart said.

The work -- both within companies and on the Internet -- is the pursuit of a much richer experience from existing data, one that will be more personal and relevant.

"Web 2.0 is about 'I want the Web to work for me. I don't want to have to work for the Web,'" Hart said. "Searching and losing two hours, five hours -- that's not efficient for me. I want information presented to me based on who I am and what I'm doing.