What exactly is big data? By the most basic definition, it is a massive amount of data that has outgrown standard analysis tools. IBM defines it with three characteristics--that is, it's large; it requires real-time analysis to reach its full potential; and it covers a variety of data types (video, click-throughs, audio, et cetera). Thus far, analysis projects have involved supercomputers that midsize businesses likely can't afford. Still, the "size" of this data is relative. You know it's big; you know it needs a lot of computing power, perhaps more than your business can afford. But what if you took a less traditional approach?

Redefining the Parameters

You don't need a supercomputer to make use of all that seemingly nonsensical data your business has inevitably accumulated--who's visiting your Facebook page and your Web site, what customers are buying, what videos people are posting about your business. Make use of the hardware you have to analyze what you can, and remember you can do some of this manually. Is it as efficient as monitoring it through a supercomputer? Of course not. You just have to narrow your sights a bit. You can tackle big data on your own scale. It won't be on the same scale as the Watson project, but wrestling that wide variety of "random" data and compiling it into something meaningful can still benefit your business.

As an IT professional at a midsize business, you probably have at least a little cushion in your budget. If feasible, start a small division. You could take one system and invest in upgrading it to the highest-grade computing level you can afford. If you can't afford to take any of your current IT team off their projects to work on it, hire a couple of data analysts for the job. Create an outline for how you will handle it, and utilize the growing edge of cloud analytics to make it more efficient. As Think IT Services details, this a great tool for midsize businesses. For example, you might want to know how many Facebook "Likes" tie into Web site click-throughs, and how those clicks tie into purchases to create a more defined image of your demographics. This is a rudimentary example.

Why Does It Matter?

You're an IT pro thinking of how you could convince your boss to approve something like that. And your boss might wonder, why bother? Your boss might think it's more trouble than it's worth. But as IBM so aptly puts it, you can answer so many questions you couldn't answer before. For a midsize business, it's all about mindset. Gather and analyze all you can with the mindset of big data: gathering larger-than-normal amounts of unrelated information and making it meaningful and advantageous.

If you expect to handle it on the same level as big businesses, you set yourself up for failure. As with so many things, if you understand your limitations, you can better utilize your capabilities. You can still make the most of it if you first define what your goals of processing all that information are. You could can build a better picture of where your business is losing money, why customers buy the products they do, and get more "in the wild", honest feedback for improvement. You can find new vulnerabilities. You can see what kinds of people are most interested in your business. You can establish new partnerships. Though your budget isn't endless, the possibilities of the kinds of questions you can answer are.

This post was written as part of the IBM for Midsize Business program, which provides midsize businesses with the tools, expertise and solutions they need to become engines of a smarter planet.