"CDA provides the governance around the data, and IT is the owner of the data," he explains. "We have business and IT practitioners working together. The CDA learns about the kinds of problems the business is trying to solve and the speed they need, so IT can build the sandbox environments that allow analytics to answer the questions quickly," he explains.
3. Stay business-focused, not data-distracted.
"Insight without action is just overhead," says Richard Thomas, CIO at Quintiles, a biopharmaceutical services organization in Research Triangle Park, N.C. That's why it is absolutely critical to keep data analytics efforts strictly focused on well-articulated and well-communicated business goals, he says.
"The whole point of big data is to create value back to the organization by operationalizing insights into action," agrees Annika Jimenez, vice president at Pivotal Software in Palo Alto, Calif., and head of the company's Pivotal Data Labs unit. "If the analytic work isn't attached directly to business drivers, that's when you see real frustration emerge."
Jimenez recommends setting up a central data and analytics team or a center of excellence with individual analytics practitioners dedicated to support different parts of the business. "Usually, we advise that people in this group sit with the business. That way, they get vetted into the business on a day-to-day basis and they bring the voice of the business back into the central analytics team," she says.
Additionally, Jimenez recommends that companies create the position of engagement manager. "These are the people who connect the dots between the business goals and the work that data practitioners are executing. They help prioritize use cases, which is a role data practitioners don't do. It's a different skill set," she explains.
At Irving, Texas-based Novation, which developed a mobile app that pushes information and alerts on changing prices of medical, surgical and pharmaceutical supplies to healthcare purchasing executives, Hari Subramanian says the best way to shift corporate culture is to adopt a collaborative approach in which the so-called "quants" and the highest-paid people work together.
"The best way is where quants work in parallel and as a team with managers to come up with a set of data that a strategist can then use [to execute a business decision]," he says. "It's usually not the quants monitoring something and then telling strategists what should be done."
4. Make analytics an all-play affair.
The marketing department might be the first department in your company to embrace analytics, but it shouldn't be the only one.
"Analytics has more impact when it becomes part of the enterprise," says Anson Asoka, vice president of global analytics and insights at Scotts Miracle-Gro. "Cutting across the organization is key. The way to connect the dots is to connect the data."
This means sharing data -- something that many top executives and traditional business managers are loath to do unless forced. This is where IT, and the CIO and CTO in particular, can be valuable catalysts, says Sanjib Sahoo, CTO at Chicago-based online trading company TradeMonster.
For example, rather than simply approving or rejecting a department's request for additional computing capability, IT can respond with data about the department's current CPU usage coupled with cash-flow information to help the requesting manager understand all options.