Hal Malchow wrote the book on what has come to be known as "micro-targeting."
The textbook-sized The New Political Targeting, published in 2003, is a dense tome written for political professionals. It is unlikely to make the bestseller lists, but if one takes its premise, that is unimportant. It only has to reach the right people.
"What most Americans and many political observers fail to understand," Malchow writes, "is that almost every competitive election is decided by a small percentage of the voters."
In major elections, fewer than 20 percent of all voters are "truly undecided," he says. The challenge is to find them — and the message that will sway them and bring them to the polls.
In essence, targeting shoots to get the magical 50 percent-plus-one needed to win.
Malchow has put his principles into practice at his firm MSHC Partners (its name recently changed from the more cumbersome Malchow Schlackman Hoppey & Cooper). In 2004, the firm provided direct mail services to federal candidates, including Democratic presidential contender Sen. John Kerry.
The micro-targeting strategy has gained popularity with the sheer amount of data — both public information and consumer information — collected in databases and available for purchase. Commercial marketers, Malchow argues, have become experts at predicting the buying habits of consumers, while political campaigns have been slow to follow. The result, he writes, is that "most campaigns do a poor job of finding the voters they need."
Since Malchow's book was published, the Republican Party has been credited with finding potential voters for President George W. Bush's 2004 re-election campaign using the micro-targeting approach that draws upon consumer research to predict political leanings.
Malchow argues for using traditional voter lists, which contain party registration and little else, and combining them with census data showing income, education level, home value and ethnicity and commercial data collected by traditional marketers — magazine subscriptions, for instance, along with automobiles and other purchases.
All that information is then analyzed by computer, using something called a "Chi Square Automatic Interaction Detector." In language only a mathematician would love, Malchow explains, "Chi Square is the statistic used to compute a probability value, or P-value. The P-value is the probability that the observed relationship between the predictor and the target efficiency could be explained by chance."
The gist, however, is that all of this analysis can lead to election success. Malchow found, for instance, that targeting people with a long travel time to work can improve turnout on Election Day, but cautions against using drive time in your strategy unless you understand the reasons behind the numbers (for example, candidates whose platforms address transportation issues).
Computer analysis can also yield a list of unregistered voters who might support your candidate. In a 2005 Campaigns & Elections magazine article, Malchow laid out a strategy to find them. For starters, traditional voter lists — those including names, addresses and party affiliations from previous elections — by definition do not list people who aren't registered. Unregistered voters, however, can be isolated by subtracting voters from the commercial databases that include every resident of an area.
The scientific, controlled nature of micro-targeting has been made possible by advances in computer processing. And MSHC Partners also has entered the online world with an Internet marketing division headed by Michael Bassick.
At a political consultant conference sponsored by Campaigns & Elections in June 2006, Bassick said the business is still in its infancy. His division, he said, doesn't make a profit, but Internet advertising holds promise. Using Internet advertising to determine who is looking at an ad and for how long, as well as micro-targeting to certain areas (such as to California residents who read The New York Times online) make banner Internet ads a cheap option for politicians.
Micro-targeting is coming into its own with the promise of Internet and cable companies that will have the potential to target very small segments of their audiences. Politicians then could reach different groups with focused messages rather than the carpet-bombing approach of buying time on the traditional broadcast networks.
For now, though, Bassick is skeptical of revolutionary rhetoric about Internet advertising, calling it "a lot of hyperbole."
"The Internet isn't going to change anything right now," he told a room full of campaign consultants at a conference this year. "There's no crazy revolution going on in terms of advertising. Anyone who tells you that is lying."