They say data is the new oil. But just like oil, data is worth much more when it’s refined than when it's in its raw state. In fact, raw data isn’t all that valuable. It’s only when that data is correlated against other data that valuable insights gets created. The challenge IT organizations face is they need to move data from one place to another to refine it and get those types of insights.
The IT industry is on the cusp of a unique moment in time when many organizations have come to understand the potential value of their data. The trouble is most of them have limited ability when it comes to turning their data into actionable intelligence.
Research released this week by Continuum Analytics, a provider of an open source platform for building data science applications, suggests that organizations are starting to approach a significant tipping point is their adoption of data science applications. A full 62 percent of the 700 business decision makers and data scientists surveyed said their organization use some form of data science on a weekly basis. But only 31 percent said their organizations were using these applications daily.
Where to refine data
The rise of both machine and deep learning algorithms should considerably accelerate that usage in the months ahead. From a technical perspective, most of these algorithms are not all that new. What has changed is the cost of aggregating all the data these algorithms need to be applied against has dropped. That cost issue has resulted in most of the usage of these algorithms being applied against data in the cloud. But it’s only a matter of time before those algorithms get applied to data residing in local data centers. In fact, just this week IBM announced its intentions to deploy algorithms against data wherever it resides.
That will prove significant because many IT organizations don’t get especially excited about moving data. In fact, from a cost and security perspective the movement of data is often viewed as the root of all enterprise evil. Most organizations prefer to bring code or machine learning algorithms to the data rather than go to the trouble and expense of moving data. Because of that data gravity issue, most raw data remains where it was first created, except for copies of that data created for backup and recovery purposes.
Improving how data becomes intelligence
What gets shared throughout an enterprise via application programming interfaces (APIs) is the analytics performed on that raw data. In effect, analytics is the refined data that is the source of actionable intelligence.
If managed service providers (MSPs) want to tap into the data-as-new-the-oil opportunity, they'll need to keenly understand how data gets turned into intelligence within any organization. After all, the organizations that refine and distribute oil in the form of, for example, gasoline, typically make significantly more money than those that merely drill for it. Unfortunately, most of the data drilling that takes place in the enterprise today still involves spreadsheets, which are essentially the digital business equivalent of a pick and shovel.