Thinking or just plain stumped?
Using Facebook to predict the yield on 10-year Greek bonds

Plus ça change – Part 1

Sometimes it feels like if you are around this industry long enough you begin to see things recycled, old ideas dressed up in new clothes and paraded out as the next big thing.  That’s how I’m beginning to feel about  big data.  I find myself increasingly reminded of the twin next big things of the 1990s: data mining and CRM.  Both of them had the power to scare the pants off market researchers by threatening two of the basic underpinnings of our industry: interviewing and privacy protection.  Both were believed to have the power to make what we do obsolete.  As far as I can tell, it didn’t happen.

The here-we-go-again aspect of big data was brought home to me while reading this article in the MIT Sloan Management Review  dealing with the obstacles to adoption of big data and analytics.  The impediments the study found are remarkably similar to those that bedeviled data miners and CRM specialists over a decade ago.  There is the same “failure” of managers to understand the power of leveraging data in different ways and the same “islands of data” problems on which enterprise-wide data management foundered. Capture copy

I suppose the major difference on this go around is that technology is a lot better, the analytic tools more powerful and the data easier to come by.  At least that’s what I hear.   But those improvements don’t solve the core problem which would seem to be the culture of the enterprise.  That’s why I don’t think big data is going to put us all out of business just yet.  Do I think that over the long-term big data will be practiced on a broad scale?  Absolutely.  I’m just not going to lose any sleep over it this week.

Part 2 will be about rediscovering some of the basic principles of questionnaire design and discovering that there is more to it than eliminating grids.