Psychographic voter profiling

Throughout the NYC mayoral campaign I kept on talking to people about how much Bloomberg was spending on “list development� and how outrageous the amount seemed. It was $5 million in May! Well the NYT ran a story earlier this week exposing what the money was going to:
[Voter] Lists like this usually include voters’ personal data – the magazines they buy, the cars they drive, their political affiliations. But as the cost of compiling Mr. Bloomberg’s list inched up toward $10 million, not even aides to President Bush, who perfected this sort of voter identification last year, could figure out where the money was going.
Mr. Bloomberg’s aides have explained the mystery: rather than trying to read the tea leaves of public records to figure out voters’ tastes and leanings, they had the money to simply call and ask about them directly. They called hundreds of thousands of New Yorkers in what top strategists in both the Republican and Democratic Parties said was one of the most ambitious pollings of an electorate ever undertaken.
They stored the answers in a vast computerized database to develop sophisticated psychological portraits of city voters – identifying eight never-before-identified voting blocs based on people’s shared everyday interests and concerns, not on their broader racial, cultural or ideological differences, aides said in interviews in the last few days.

Now, the big question I have, neither asked nor answered in the article is: what happens to the data? Bloomberg has built $10 million worth of advanced demographic and psychographic data, and he can’t run for Mayor again. He is not going to run for either of the other two citywide offices, and has chilly relations with other city and state Republicans. Essentially the data is going in the garbage, or will age beyond utility. All that will remain is the methodology. And using the methodology is expensive! In this case $10 million to model a 3.5 million voter electorate ($2.85 per voter) and only 700k of which voted ($14 per vote). If you translate that to national numbers, that is something like $1.54 billion ($14 per voter x 110 million voters). Though you would only have to use the methodology in competitive swing states, you are still talking about 10s of millions of dollars to built these profiles. By comparison, Bush spent less than $5 million on voter data in 2004.