Big data is top of mind for the C-suite and IT executives today, particularly in the realm of marketing data and marketing analytics, says Manjit Singh, CIO of The Clorox Company. While organizations are focused on big data, many are failing to act upon their data because they simply get sidetracked.
Where the data challenge lies
"The challenge is not in collecting the data," Singh says. "We're challenged in how to get insight out of the data — what questions to ask and how to use the data to predict results in the business."
Singh is no stranger to big companies with big data. Prior to Clorox, he served as CIO of the Las Vegas Sands and Chiquita International, and as the Asia Pacific regional CIO for Gillette.
[ Related: 5 big data technology predictions for 2015 ]
For Singh, the answer is to always focus on enabling strategic business objectives, to break things down into discrete tasks with measurable outcomes.
Prat Moghe, CEO of Boston-based startup Cazena, which has brought together veterans of the Netezza team to help big companies focus on big data applications rather than infrastructure, agrees.
Cazena CEO Prat Moghe and Manjit Singh, CIO of The Clorox Company, discuss the CIO's view of big data and cloud transformation.
"The goal of big data is not better data science," Moghe says. "The goal is to be able to leverage data to achieve business objectives. Too often this idea gets overshadowed by our tendency to focus on new tools and technologies rather than business outcomes."
Chase entrepreneurial spirit, not scientists
Rather than chasing hot new tools and hiring large teams of data scientists to set up the technology and manage the data, Moghe argues that CIOs should focus on finding employees with an entrepreneurial mindset that can drive agility with big data.
"By doing so, big data will become the bridge that helps the CIO role break out of the organizational silo," he says.
[ Related: 4 ways to beat the Big Data talent shortage ]
When you start from the business use case, Singh adds, infrastructure questions become much easier to answer.
"One of the best examples is looking at your volume shipment data and connecting it to certain initiatives you have in the business, like sales," he says.
For instance, you may want to measure the effect of a promotion effort. But maybe there was a snowstorm in the region during the period you're evaluating. If the promotion didn't meet expectations, was that due to some quality of the promotion, or was the weather to blame? You need to bring in weather data for that, but you don't need to know what the global weather was in that period, even if you have access to the data.
You don't need a weatherman to know which way the wind blows
"You want to look at weather patterns that are locally relevant," he says. "You want to figure out how to narrow down the context of the data source so it's yielding a relevant result for you."
In other words, for any given business objective, you probably have a lot of data elements that are nice to know, but that are not necessary to achieve that objective. To move the needle in an efficient, cost effective and agile manner, you need to carefully narrow down the data to that which is actually useful in a given business context.
As an added benefit, keeping the context narrow can help you identify the actual causation of events.
"That is harder to do than it sounds," he says. "We all like to think in broad strokes. Every initiative is intended to have a business result. But what is the actual business result? What did I do to get there and how do I determine that what I did caused the effect?"
To do that, Singh says, you have to decompose a problem to the correct elements so you can measure it appropriately.
"It's more of an agile mindset," he adds. "Test and learn. Try something and see if it moves the needle correctly. If it does, great. If not, try a different approach."
Cloud adds agility to analytics
Singh notes that when it comes to infrastructure, the issue always comes down to money. There is a mismatch between the amount of money the business has available and ready to spend on infrastructure and how quickly the business wants to get an analytical capability ready to produce a business result.
That's where cloud comes in, he says. You don't necessarily need to build out your own infrastructure. Instead, you can tap into someone else's capabilities, spread out across multiple companies.
"Software-as-a-service and cloud has given us back the ability to test and learn without making a massive investment in software licensing and infrastructure spend," he says.