Sounding Board: The numbers game

Both presidential campaigns were driven by data. What can retailers take away from this past election?

November 6, 2012 will go down as a seminal date in American history.

Yes, the country’s first African-American president was re-elected commander-in-chief, the first Mormon candidate was a hair’s breadth away from the White House and the entire electoral process was another triumph for democracy.

My interest is far less lofty and apolitical.

To me, 2012 was the year data mining and analytics really came of age. Some even say it was responsible for electing one man president over another. Therein lies the overarching lesson for retailers—do not assume you know your customers (i.e. voters) or you may find yourself looking for another job.

Polling by politicians is certainly nothing new. It was kicked up a few notches by the Obama campaign in 2008 and became a must-have science this past year by both candidates, albeit used somewhat differently. In fact, Democrats and Republicans spent in excess of $13 billion on data acquisition.

Mitt Romney used it to pinpoint big donors who would give $2,500 or more—even in Democratic strongholds—and he was pretty successful in doing that. But Barack Obama used it more effectively in getting his message to the right people. There are those inside and outside the Washington Beltway who are convinced that who he reached was more important than what he said and that cutting-edge analytics won him the election.

The nerve center of what was dubbed  “Project Dreamcatcher” was “The Cave” at Obama’s campaign headquarters in Chicago. It was populated by an army of analysts and number crunchers using secret code names to delve into the habits of different groups. It has something of a Big Brother-ish quality that Democratic strategists would rather not discuss.

The whole thing was overseen by Obama’s “chief scientist” Rayid Ghani, the guy who held a similar title at Accenture, where he was expert in developing analytics to study the consumer behavior in different categories and the impact of personalized supermarket sales promotions. Who said politicians are people, not products?

Basically, Ghani’s algorithms were predicting what people would do.

The analogies between what went on behind the scenes in this year’s presidential race and everyday challenges among supermarkets are pretty obvious:

  • Do not leave money sitting around unspent;
  • Get your message across by analyzing where to buy ads; and
  • Focus on undecided voters (shoppers) who often make up their minds on what to buy (who to vote for) at the store (in the voting booth).

Some say I am preaching to the choir. But you would be surprised at how many retailers continue to overlook the real value of customer analytics. The recent recession, whose shadow still lingers over the U.S. economy, will still have an impact on IT budgets in 2013 and cost cutting is still very much a mandate among chains and independents.

But customer analytics is not simply an expense on the balance sheet. Properly used, it is an asset and a cost saving for anyone planning to stay in business and keep customers from defecting to competitors.  There is a clear return on investment that can be measured in terms of increased sales and profits.

There are, of course, pitfalls. Many groups remain concerned about gaining extensive customer information that violates a customer’s right to privacy. The solution is to explain to customers what data they are collecting and what it is being used for. You have to prove why it is in their best interests to provide data for tailored service and products developed especially for them, or for marketing campaigns targeted only where appropriate.

One possible solution is being studied by Hector Garcia-Molina, a professor at Stanford University. It would keep databases from extracting too much personal information, but still be able to adequately slice and dice the data.

Additionally, sources strongly suggest not letting the IT department hold sway over consumer data. Rather it should be a collaborative effort between IT, marketing and merchandising, enabling retailers to turn raw data into useful information.

The endgame is there is no such thing as too much data. Just ask the guy at 1600 Pennsylvania Ave.  I do not have his number, but he is in the book. 

This entry was posted in 2013 01 Article Archives and tagged , , . Bookmark the permalink.