Thomson turns to XML for report indexing

Thomson Financial is rolling out XML tagging software that will automate the indexing of investment analyst reports as part of an...

Thomson Financial is rolling out XML tagging software that will automate the indexing of investment analyst reports as part of an effort aimed at giving customers greater search capabilities.

The financial information provider is expecting resulting cost savings and improved sales to generate an 18-month return on investment.

Every month, New York-based Thomson has been manually indexing 40,000 PDF financial reports about various companies and industries. Thomson then posts the information on a secured Web site to sell to the financial services industry.

Thomson hired about 15 data indexing specialists to "crudely index" 120 documents each day by hand so customers can search for specific reports, according to Jeffrey Mastendino, vice-president of content management at Thomson Financial. It took about six months to train the specialists, and the process was time-consuming, labour-intensive and often inconsistent, he said.

In December, Thomson and New York-based ClearForest signed a deal worth "hundreds of thousand of dollars" that calls for Thomson to use ClearForest XML-tagging software that will automatically set up research documents for archiving and offer more precise searches on data, according to Mastendino.

The initial ClearForest implementation should be completed this quarter and deliver some heady returns, said Mastendino. "We think on the report-level side it's going to be pretty significant to savings. Turnaround time will be almost immediate for the bulk of documents," he said. "The implementation we designed will be able to handle that systematically, and probably two-thirds of the research will be able to be handled automatically."

Thomson is part of a growing number of financial services companies to jump on the XML bandwagon. "In financial services, there's a lot of interest in this type of technology," said Laura Ramos, an analyst at Giga Information Group. "In some respects, the economy really put a lid on the budget to pay for it," she added.

Some financial services firms are doing more than experimenting with the development schema. Metropolitan Life Insurance is building a realtime relational database based on XML and Fidelity Investments recently completed an enormous retrofit of its corporate data to an XML format.

Read more on Business applications