Franks, chief analytics officer at Teradata, is like the chief software
architect at Microsoft, the vice president of search at Google or the head of
design at Apple. He’s a voice of authority in his discipline in a way having
little to do with his position on the organization chart.
Franks spoke today to the Business
Marketing Association of Atlanta, describing the practical challenges of Big
Data. While Gartner says Big
Data has entered the trough of disillusionment, Franks remains enchanted.
There’s a bubble in the market for Big
Data, Franks conceded, but the stock market says nothing about the long-term
power of analytics to reveal practical money-making connections.
Take mixed martial arts and the History
Channel. Seems like an odd match, unless you’re looking for the physical
equivalent of Epic Rap
Battles of History. But it turns out that Franks' colleague noticed a connection
between the two media properties while looking for ways to boost pay-per-view
advertising value for a UFC match. “It wasn’t just generic History Channel,” Franks
said. “It was one show on the History Channel. Swamp People.”
The gold nugget in the data: UFC could
direct some of their very expensive ESPN advertising budget to Swamp People at
more value for the dollar, because advertising on the History Channel comes
Big Data disrupts industries in much the
same way the Internet does, Franks said. With the History Channel, for example,
Big Data exposes the relative lack of rigor used by broadcasters in traditional
media measurement. Folks selling ads don’t know what they’re worth. They’re
guessing. But a Big Data professional will have a much better guess.
For competitive intelligence
professionals, Big Data represents a major opportunity. Franks figuratively
encourages firms to “ignore
their own business” and find outside companies that might profit from the
valuable data businesses collect. Firms have tremendous information about their
own processes and their markets, but they’re also unintentionally gathering
huge amounts of information about the competitive environment. Some of that
will be about their own competitors – that’s not data they’ll share. But about
firms they don’t compete with?
One might imagine brokering access to
aggregated sales data from a credit card company, not just of one’s own
merchandise but that of a competitors as well. Or perhaps an analysis of
traffic patterns from a telecommunications provider can show how a rival shop’s
customers come and go.
Time to cut a deal.
One drawback to all of this – it remains
labor intensive. To start, data often has to be “cleaned” of errors, a task
that might take 75 percent of a project’s time, Franks said. And there’s a
major skill shortage in analytics to boot, he added.
“That’s why it’s great to be me.”