Everyone’s talking about big data these days, but few have figured out how to transform all that information into insight.
A McKinsey & Co. survey last year confirmed many companies — particularly digital non-natives — haven’t yet achieved desired results or significant returns on investments from big data initiatives.
Average initial increases in revenues after big data investments were a meager six percent.
To be sure, big data is about more than simply tapping into the so-called 3Vs (variety, velocity and volume) of data. There’s a need to take more of a big picture and holistic view of how data can lead to value creation, says McKinsey & Company partner Jeff Jacobs.
How to get started with big data analysis? We suggest a four-fold model.
Build the right framework for data analysis
Big data doesn’t happen without the right IT systems and software.
While it’s possible to simply plug databases and stand-alone analytics tools into legacy systems, the real value comes from building integrated data platforms in the cloud, and tying them into legacy systems, says Scott Schlesinger, information technology consultant for Ernst & Young.
This allows an enterprise to connect disparate data sources, aggregate data on the fly and run analytics in real time. It breaks down data silos.
We expect the CIO’s team will govern the IT setup. But marketers can influence it by clearly articulating the desired outcomes; that the net-net should be an improved ability to reach consumers with highly personal and contextually relevant promotions.
Look across platforms
Making data work is about combining information in ways that illuminate previously hidden patterns.
Athletic apparel and footwear retailer Finish Line took this approach toward its email newsletter program. By combining data from point-of-sale systems, social streams and other sources, the company boosted open rates by 50 percent and improved conversions. The insights also drove a 30 percent increase in gross return on ad spend for Facebook.
“Our goal is really just making sure that we’re putting the right information in front of the customer, and then creating what we call the frictionless, omnichannel experience,” says Stephanie Bleymaier, director of digital personalization and loyalty for Finish Line.
Look across departments, too
The companies that most-successfully harvest big data are the ones that build a framework for communication and collaboration across the enterprise.
This means busting through silos to enable data sharing within the organization and out to partners. This creates an ability to slice, dice and understand data — as well as organizational needs — more broadly.
Beyond open rates and conversion rates, Finish Line has driven overall improvements in marketing and sales by ensuring everyone views and works with the same data, Bleymaier says.
“Data has to be identified, acquired, organized, filtered and cleansed, then integrated and stored before it offers real value to the end business consumer,” Schlesinger says.
Test and learn
With the right IT framework and approach in place, the fun can begin — A/B tests and other marketing trials. It’s about piloting, testing and conducting proof of concepts to learn what works, Schlesinger says.
Good news: the McKinsey study suggested that over the span of five years, most companies see modest increases in returns on investments in big data.
Not-so-good news: the increases were only from 6 percent to 9 percent, overall.
Proof positive that the companies learning how to turn information into insight remain far ahead of the pack.