Data Science: Get way smarter – or else
August 20, 2013 | Alicia Androich | Comments
Data scientists can tell you everything from which customer you’re about to lose to how to best price your products. Using their unique and coveted skill set, they deftly mine golden nuggets from reams of data. A look at who they are and how they work their data-driven magic to help marketers
Christopher Berry is the co-founder and chief science officer at Authintic, an analytics technology company that specializes in social segmentation and understanding consumer brand preferences. He has seen some cynicism around big data, but is determined to use new technologies to create better user experiences for brands.
What’s your simplified definition of big data?
“Data that does not fit on a single computer.”
What’s your definition of a data scientist?
“An individual that combines three skill sets: they understand the business, code and statistics. [Data science is] at the intersection of those three.”
What got you interested in this field?
“My graduate degree is in applied statistics to public policy. For me, hardcore statistics has always been a means to affect some sort of change.”
Does it feel like it’s still early days in terms of how Canadian marketers are leveraging big data?
“The future is already here, it’s just distributed unevenly… I think very few marketers in Canada would turn down a discussion about the applications of big data because they want to do as much research and know what’s going on as possible. But whether or not the cash or the budget line items are there to begin piloting or executing really, really depends. [For some] sectors of the economy it’s an absolute mandate that they have to get way smarter with data science or they’re going to be facing severe disadvantages in the future. Certainly airlines are aware of creeping incremental competitive pressures, and with the entry of U.S. firms into telco… and increasingly a lot of e-commerce leaders in the country are aware that they have to step it up.”
How do you compare traditional database marketing to what big data is today?
Churn analysis and the models that go along with that—like shop ’til you drop models and various types of hidden Markov models that can be used to predict churn – can help managers build business rules around intercepting people that are likely to churn. To me, that’s traditional database marketing. There’s a movement amongst business intelligence people that their common refrain is “There’s nothing new about big data; we’ve been doing churn analysis since the ’60s.” They’re right if we’re talking about those base models. What I think might be new is if you’re, with permission, observing a [smartphone user] actively engaged in a comparison of exclusive handsets that are only available by other carriers, for instance. The incremental new data sources that are leveragable because you can access them using application programming interfaces might enable a better model. If you have a better model and you can get 1% or 2% better performance that can translate to millions of dollars for a firm every single year, that might be an evolution of existing business intelligence methodologies.
And if you want up-to-the-minute news coverage of marketing, media and PR news, subscribe to our twice-a-day free e-newsletters.