Big data is not the new oil, but data science can make it look “oily”
Virginia Rometty is President and CEO of IBM, one of the largest technology consulting companies in the world. Her voice has significant weight in the corporate tech world, with Forbes’ magazine recently placing her among the “World’s 100 Most Powerful People”. A few years ago, she famously stated: “Big data is the new oil.” A statement from such a credible source should bear some truth, shouldn’t it? If stock markets are to reflect projection of the future, Rometty’s statement can only be confirmed: Big data is indeed the new oil. At least if one assumes companies like Google and Facebook as representatives of big data business models. Both rank in the Top10 by market capitalization with significant double-digit growth in the last year. Oil companies like Royal Dutch Shell and Exxon Mobil have long been surpassed by the two tech giants. Nevertheless, while big data might start to unfold a similar impact on the economy as oil did (and continues to do), it is not the new oil. Here is why.
Data vs Oil: Resource discovery
Oil is a scarce resource. Theories about peak oil have long been established, even though drilling and sourcing are still on high levels. Nevertheless, oil prices have been plummeting recently, reaching record lows at $38 and below. With scarcity as one of its main value components, the rise of competitive renewable energies as well as alternative drilling technologies (“fracking”) have led to cheers on the consumer side (e.g. drivers) and sorrow for oil executives. All of the sudden, substitutes are available and diminish the perceived value of oil for our economy. With data, it’s a different story. The more data we have, the more value we can generate. More data can just enhance the quality of analysis, not challenge the value of existing data. With that said, with new technologies, sensors and the like, data sources will increase over time. A “peak big data” is not around the corner.
Data vs Oil: Refinement
While oil and data are fundamentally different as a resource, they share a similarity regarding the need for refinement. Crude oil itself is of very little use for industrial applications. However, refined into gasoline it enables urban and commercial transportation. Similarly, plastics, chemicals and the many manufacturing plants all depend on oil. In a raw state, data is also of no use for corporations and businesses.
With that said, refinement is fairly easier if you deal with a more or less standardized resource such as crude oil. With data, things look different. It comes in all kind of formats, with all kinds of quality attached. One of the main activities in terms of “refinement” is precisely this standardization in order to generate insights across domains and sources. Big data is heterogenous in nature and the challenge is to still utilize it for analysis, prediction and forecasts. Oil, on the other hand, is quite homogenous and easier to process. Consequently, ownership of oil is far more valuable than the capability to process it. While ownership of data is important, the capability to standardize and analyze it for insights is the key value driver.
Data vs Oil: Value Creation
The starkest difference between oil and big data, though, stems from its value creation. There is a very clear use-case for oil in today’s economy. Most prominently, refined oil, especially as gasoline, drives value creation for consumers in the world’s combustion engines. There is a very standardized set of applications, which heavily depend on this commodity. For data, once again, this does not apply. On the one side, this makes big data incredibly powerful as there are millions of use cases. But also incredibly useless, without a tailored solution to each use-case. And it is precisely the necessity for tailored solutions, which make big data more valuable for those, who have a team of capable data scientists readily available.
Within the data value chain, the analytical capabilities of those who make sense of big data are the key ingredient. Thus, big data is still not the backbone of the economy as it is too volatile, too fragmented and insights are dependent on non-transferrable intellectual capability. However, it – already today – enables the economy to work smarter and more efficiently. Even oil refineries are relying on big data to streamline production and output.
Data science can make big data look “oily”
The limitless applications of big data are its main challenge. While big data may have an increasingly important role for our economy, just like oil had for the last decades, big data will never be the new oil. At the moment, efforts to extract value from big data are often too burdensome. Analytical value is thus hampered by up-front costs. Nevertheless, data science can help to extract as many oil-like features of big data in order to finally lift its full potential for the economy. With increasing efforts to aggregate and standardize data sources, more value potential will be tapped upon. Big data, in the end, can be quite oily.