Big versus other data

If numbers can make you look smart, it should stand to reason that big numbers can make you look super smart.


Big and small: each has its place.

But then, looks can deceive.

Don’t get me wrong. I hardly need enumerate the virtues of big data. It reveals behaviors, preferences, and predilections on a mass scale. All of which are useful.

But note the “mass” in “mass scale.” Then, recall that a strong brand is, by design, not for everyone. It is important to guard against letting big data homogenize at the expense of niche opportunities, and against letting it keep you from winning by zigging while the rest of the world zags. Unless yours is a financial institution like all others, do not let big data lull you into assuming that your clients are like all others.

While big data is useful for revealing how we behave, it is less so for predicting how, given the chance, we might behave. Big data did not predict, nor could it have, that consumers would willfully stand in line to overpay for coffee, text more than talk, or, for that matter, buy shoes—something that data revealed people had to buy in-person—online.

That is why wise marketers take care to round out big data with other, smaller, more specific data. And to test small before going big.

On the technical side, beware applying the same logic to big as to smaller data. Take, for instance, p-value, defined (loosely) as the probability that the null hypothesis is wrong. The bigger the data, the greater the danger of false positives. As Minitab’s Patrick Runkel explains, a too-large sample can appear to magnify numbers you would rightly ignore in a smaller sample. And as neurologist Steven Novella explains, “The p-value tells us the probability of the data given the null hypothesis, but what we really want to know is the probability of the hypothesis given the data. We can’t reverse the logic of p-values simply because we want to.”

Big data can offer an alluring but dangerous out for the lazy marketer. (Perish the thought that lazy marketers rank among this blog’s readers. I’m taking about, you know, those other marketers.) Embrace it. Stay current with it. Mine it for every gem. But don’t assume the thinking has been done.

Comments are closed.