Citi Exec on the Culmination of Two Centuries of Data Quality Management

At Citigroup, data quality isn’t just an issue that has emerged in recent years: it’s a 200 year-old problem requiring ongoing maintenance programs that begin again as soon as a current project is finished, according to officials who spoke at last week’s Waters USA conference.

“Citi started in 1812, and its data issues started in 1813,” said Don Callahan, global head of operations and technology at Citi, which still maintains historical data as far back as 1812, including early shipping records, he said. “We used an ‘adjacency model, which means we didn’t integrate data. People kept their cards close because that information is power—and that [approach] is bad within an organization, because shared information is more powerful.”

Though data management is an age-old problem for the bank, recent market events have given data management projects renewed impetus and have helped data professionals make their case for enhanced data quality programs.

“The financial crisis gave us the opportunity to take a step back and come up with a common data architecture,” across all the companies that have merged together to create the present-day Citi, Callahan said. “We spent a significant amount of money and time on this: I presented the board with a five-year roadmap to take 23 companies and unify them on common data.”

And—just like painting the Golden Gate Bridge—as soon as Callahan’s team finished that project, they had to start over again, and will have to continue doing so to match the ongoing evolution and constant acquisition of few data.

One of the challenges was changing the mindset of allowing different businesses to keep data to themselves. “Some of it was having the discussion with business leaders, and saying, ‘What you thought was your data is our data,’” Callahan said.

But once the firm can establish common ownership and ensure that technology exists to move data from point A to point B anywhere within an organization without recoding at each point, the benefits are higher levels of insight firm-wide.

“I’ve been in the industry for 40 years, and I can tell you that without solid data, you won’t be able to build any kind of organization. You turn data into information, information into knowledge, and knowledge into insight, and that’s when you get paid, whether it’s a trade, a credit card transaction, or a decision about a mortgage. But it starts with data, and if that data is flawed, then the outcome is flawed. So data is not just the fuel of growth, but is essential to any organization,” Callahan said.

It’s during this transformation process that technology has an important role to play. “Technology was our friend,” Callahan said, adding that he was surprised how easy technology made it to move data from point A to B.

Christian Nentwich, CEO of data management platform vendor Duco, noted two contributing trends: first, that the cycle time for displacing older technologies is getting shorter; and second, these new technologies are enabling firms to “empower their data citizens… and put more power at the fingertips of a company’s employees. “There is a lot of potential to use AI [to perform] this first step of turning data into information. We are doing a lot of work on that, and are looking to roll something out next year,” Nentwich added.