On Topic: Big Data can be business' friend

Ready or not, business world, here comes Big Data.

That's the message this month in both Foreign Affairs and McKinsey Quarterly, publications whose writers generally seem to know a lot about a lot of things (even if you don't always agree with their conclusions).

And these articles comes with pretty hefty warnings.

If you don't know what big data is, well, that's part of their point. We all need to understand it, what it can do - and what it cannot - and then come up with a plan to harness it for our companies.

First, a definition, courtesy IBM.com: Humans on planet Earth generate some 2.5 quintillion - in America, that's 1 followed by 18 zeros - bytes of data every single day. Ninety percent of data available today was created in just the past two years.

These details have been gathered from everywhere and includes everything flying about in our information era - social media chatter, climate sensors, medical and Internal Revenue Service records, property-tax payments and water bills, Homeland Security statistics, traffic camera video feeds, cellphone signals, car rentals, fishing licenses … . Everything, from everywhere.

For hundred of years statisticians happily worked with the tiniest nuggets of information, Kenneth Cukier and Viktor Mayer-Schoenberger explain in their Foreign Affairs article (adapted from their new book, “Big Data”). This is called sampling.

From those winnowed-down bits of facts - think of exit interviews on election day, the authors suggest - you successfully could extrapolate a winner.

Sometimes. Maybe.

But what if - given the miracle of modern computing - you could gather an excessively, seductively massive amount of information? Practically everything?

You're no longer panning for gold, you're driving an earthmover.

One outcome is things will get powerfully “messy,” the authors note. With so much data, you'll end up netting a bigger number of inaccuracies.

The thinking, however, is the benefits of that huge amount of data will outweigh the untrustworthy stuff.

And then, oh, the places you'll go: By pooling all this information and sifting for emerging trends, proponents argue, retailers could predict what buyers might want even before those potential customers realize it themselves. Health care professionals could pick up on red flags of an oncoming epidemic.

Meanwhile, writers at McKinsey Quarterly note the case for using big data, along with advanced-analytics management, “is no longer in doubt. … (Their use) can deliver productivity and profit gains that are 5 to 6 percent higher than those of the competition.”

But you'll need a strategy to focus on determining what you want to learn, how to organize and analyze it, and how much you want to spend to accomplish all this, they warn.

Otherwise, like the pharaoh's army of the Exodus, you'll find yourself dangerously underwater in pretty short order.

And there's this: Big data is just that - data. It's another tool, and it comes without intuition or serendipity.

A person still needs to examine the stuff and determine, “Ah, I see by choosing to do A and after considering B, this is the next likely step.”

That eureka part has to come from us.