A Queensland government department is saving more than $300,000 a year in known and measurable costs because it addressed the quality of data in one of its smallest transactional procedures.

Janet Brimson, head of the information management practice at independent IT services consultancy iFocus, says double handling of data alone was costing the department $72,800 a year. Improved granularity of data for reporting is saving $185,000, while the agency is also saving an unknown amount through elimination of errors.

Brimson was speaking after warning at last month's Strategic Data Quality Management Conference in Sydney that neglect of data quality management can cost government organisations millions in lost revenue - as well as to repair - yet it is very rarely included in IT budgets. Brimson told CIO Government organisations that ignore data management will be hit by a data "domino effect". "Data that is incorrect at the source travels across all systems and applications wreaking havoc," she says. "This will worsen until the source data is located and fixed.

"I have one federal commission currently that ends up with hundreds of duplicate records in invitation lists for major events that it needs to coordinate. They are losing money on excess printing costs as well as three to four days per event to hand-correct the data. Then many of the invitations are returned as address unknown, and they spend several more days trying to locate the individual prior to the close off of numbers. It blows the budget out for the events management continually across several states, so they are putting in data quality practices to prevent the losses," Brimson says.

Many agencies across all levels of government are suffering continual erosion to revenue because the names and addresses of companies in their CRM systems do not match up, making it impossible to consolidate reporting, she says. Consider the many permutations of Coles Myer alone: sometimes it is recorded as Coles, sometimes as Myer, other times as Coles Myer or Coles Myer Pty Ltd or Liquorland.

"These permutations are based on correct spelling only and don't take into account the misspellings that can be in systems. From a business intelligence perspective it can take days to consolidate and match the information prior to reporting, weeks to find and correct mistakes and then most of these organisations have found that they still didn't get to interrogate all the data [because] there were other repositories that were not declared to them," Brimson says.

When it comes to IT budgets, data quality management is not factored in, which ultimately costs organizations millions of dollars in lost revenue.

"IT budgets are overwhelmingly oriented towards new technology purchases and implementations. The business cases rarely take into account the cost of information creation and management that will feed the new applications," Brimson says.

"Information is a corporate business asset. Maintaining the information that an organization already has is very rarely allocated sufficient or any funding, but when information integrity is eroded it can be difficult and expensive to fix."

Brimson says organizations today run so many systems and have so much information that data housekeeping can no longer be ignored.

"A "housekeeping" budget should be included as part of any IT budget if companies and organizations are to avoid being swamped with data problems in the future," she says.

"Data quality and the management of information throughout its lifecycle is fundamental best practice."