As much as 80% of business data today resides on legacy mainframe systems, according to Dain Rauscher Wessels, a research firm. Efforts to build Web-enabled procurement, order-processing, and customer-service systems require integrating that data with new E-commerce or customer-relationship management applications and data warehouses.

Before that integration work can begin, however, IT managers must study the legacy data to uncover hidden quality problems and inconsistencies. A bank trying to pull together information from multiple databases to create a single view of its customers, for example, may discover that each database uses a different address format, which can make it difficult to combine the data.

The most significant feature of the Axio technology is that it examines the granular, or detailed, data in a legacy system, rather than simply analyzing the metadata as most data analysis tools do. "This product goes through the data column by column, row by row, to figure out what's in there. I don't know of any other product that does that," says Michael Reed, data warehouse and enterprise application integration research director at Technology Evaluation.com Corp.

Toyota Motor Sales USA Inc. in Torrance, Calif., began using an early release of Axio in February as part of a project to rebuild hundreds of legacy databases--some of them 20 years old--across the company's operations and divisions. The goal is to understand the type and quality of the data the systems contain, says data-quality analyst Jim Lovinggood. "But we're also capturing that data at the enterprise level in a repository for our [planned] data warehouse," he says.

The Axio suite includes a repository, a data analysis and profiling tool, software for designing staging and target databases, a repository navigator, an import tool for extracting relational and legacy data, and a source profiler to manage data preparation. The suite starts at $1 million for five users.