IBM Big Data expert shares tips with CPAs

By Ken Tysiac

Paul Zikopoulos
Paul Zikopoulos at the AICPA spring Council meeting in New Orleans (Photo by Cheryl Gerber)

Paul Zikopoulos uses the mall as an example of how Big Data can be used for meaningful business insights.

When you go to the mall, who doesn’t sign onto the “free” wireless internet service so you can use your smartphone and keep data costs under control? Because you have likely registered as part of a mall loyalty program (making you eligible for real-time offers while you shop), you’re on record as a 42-year-old woman with two children who loves shoes and chocolate chip cookies, alongside other attributes.

Through that “free” wireless connection, your movements are tracked around the mall, along with the travels of other customers who use the free Wi-Fi to enhance their shopping experience.

The mall isn’t giving you free Wi-Fi because it doesn’t care about costs—it is learning. For example, when a particular area shows up as a hot spot with high traffic for female customers, the mall can ask for premium rent from women’s apparel stores in that location.

“I want people to imagine the art of the possible,” Zikopoulos said, “and apply some of the ideas to their own businesses.”

Zikopoulos, the IBM vice president for competitive and product strategy for Big Data and Analytics, spoke Tuesday at the AICPA spring Council meeting in New Orleans about opportunities that exist through the use of data analytics.

During an interview last week, he described the advantages of using Big Data to inform business decisions.

He said organizations that want to start using data analytics to gain an edge can follow a few important steps:

Commit to a culture change. Projects may fail to get off the ground without acceptance within the organization of the concept of using Big Data.

“You have to decide as a company that you want to be data-driven,” Zikopoulos said. “… You’re on a journey to become a data-driven and data-led organization.”

Build governance around data. This includes gaining an understanding of the veracity of the data and addressing security and privacy.

“I recommend that clients start with governance,” Zikopoulos said. “… I believe governance should be put upfront.”

Don’t set out to hit home runs. Chances are that when you start using Big Data to inform decisions, if you start with the premise that your first efforts have to result in a home run that revolutionizes your entire business model or an industry, you aren’t going to hit a home run.

“You just have to get started,” Zikopoulos said. “Get to first base. Get on base.” When you try to hit a home run, you may strike out, so it’s best to start small.

Let the business drive the projects. Start with the problems you want to solve, and then find ways to use data to solve them. If you start with a scientific data analysis without a clear business imperative, your project may end up with no useful application.

For example, a company may start with an objective of getting better control over inventory, or by developing an understanding of how weather patterns may affect purchasing behavior.

“I always say, lead with a line of business,” Zikopoulos said.

The confluence of social media platforms, mobile technology, and cloud platforms has given organizations the ability to derive many different insights from data, Zikopoulos said. He said now is the time for businesses to start thinking about how data analytics can work for them.

“It’s going to change the way that we transact and the way we interact,” Zikopoulos said. “It changes the behavior of the new buyer and their expectations. … And if all that change is present or on the horizon, then I don’t think I need to tell anyone who wants to be a leader in their field that they need to be looking at this.”

Ken Tysiac ( is a JofA editorial director.

Where to find June’s flipbook issue

The Journal of Accountancy is now completely digital. 





Leases standard: Tackling implementation — and beyond

The new accounting standard provides greater transparency but requires wide-ranging data gathering. Learn more by downloading this comprehensive report.