"Unknown unknowns" are going to form a significant part of the "rosy" future for data science and data scientists.
That's according to Giles Pavey, head of data science and innovation for dunnhumy, an analytics consultancy whose clients include the likes of Tesco, Coca-Cola and Unilever, some of the biggest companies in the world.
"A big part of where we see things going is data scientists, rather than solely creating reports or data cubes, actually creating data products," Pavey told Computing, when asked about the future of data science during an interview at the SAS Premier Business Leadership Conference in Amsterdam.
Referencing perhaps the best known speech made by former US secretary of defense Donald Rumsfeld, Pavey said that analysing "unknown unknowns" is going to become an increasingly active area of data science.
"In terms of experimentation, there's the Donald Rumsfeld quote about the ‘unknown unknowns'. I think that's a key area for data science," he told Computing, arguing that currently data scientists are used to examining how combinations of known factors have affected businesses.
"A lot of business intelligence has historically been the business asking analysts: 'Our sales went down because of the horse meat scare - can you tell me what happened?' for example. That's a known unknown, whereas it should be about just getting in among the data and seeing what patterns and recommendations we can find," he explained.
Pavey went on to say that big data needs to be seen as a way of generating revenue.
"Ultimately it's about moving analysis and data from being a cost to a key revenue driver. Fortunately for dunnhumby that's something we've always known about because that's what our business does, but obviously the vast majority of businesses aren't analytic consultancies," Pavey said.
Big data was a much-discussed topic at the conference, with Mikael Hagstrom, executive vice president of SAS Europe, Middle East, Africa and Asia Pacific, labelling it as "absolutely critical" to boosting the European economy.
By eliminating high entry costs for big data analysis, you can convert more raw data into valuable business insight.
A discussion of the "risk perception gap", its implications and how it can be closed