New research from Harvard Business School shows that mass advertising is better at swaying undecided consumers while face-to-face personal selling is more suited at closing the deal for those already leaning toward a particular product.
In this case, the products in question were the Democratic and Republican presidential candidates, but the results could help many marketing teams decide how to allocate scarce resources between mass advertising and personal selling efforts, or, as the researchers call it, between the air war and the ground game.
“The reason Obama won was because of the utilization of ground forces—the personal selling and get-out-the-vote strategy.”
"I think this research connects beautifully with a normal business operation within a corporation," says Doug J. Chung, an assistant professor in the Marketing unit at Harvard Business School.
The research, reported in the October working paper The Air War versus The Ground Game: An Analysis of Multi-Channel Marketing in US Presidential Elections, was cowritten with HBS doctoral student Lingling Zhang.
Chung and Zhang pored over 18,650 observations on voting outcomes and campaign activities for the 2004, 2008, and 2012 presidential races. They studied the number of votes cast in each county for the candidates and used registered party affiliation at the county level to look at how campaign effects differed depending on the level of voter partisanship.
The researchers hoped to answer several questions. How do different types of advertising—candidates' own ads versus outside ads—and personal selling from field operations—in the form of door-to-door visits and phone calls to voters—affect voter preferences? And how do these campaign activities affect the outcome of elections through their diverse effects on various types of people?
Candidates have been investing heavily in both mass advertising and personal selling. For the 2012 election, for instance, the Democratic and Republican candidates, together with their allies, spent over $2 billion combined—the most expensive election in US history. Candidates have not only advertised more on TV and in other media, but also upped the ante on personal selling efforts with an increased number of local field operation offices as a way to reach out to voters.
But which strategy was more effective for the election of a president, and why?
What Chung and Zhang discovered was personal selling—the ground war—had a stronger effect on partisan voters, but a candidate's own advertising was better received by nonpartisans.
Personal selling accounted for Barack Obama's victories in the 2008 and 2012 elections. Advertising was critical only in a close election—George W. Bush's (HBS MBA '75) victory in 2004. There was no one-size-fits-all marketing approach that swayed all voters.
"The reason Obama won was because of the utilization of ground forces—the personal selling and get-out-the-vote strategy," Chung says. "The idea was not to change people's minds who didn't like him, but to get people who liked him to go out and vote for him. It was a 'we need you' message—we need you to come out and vote."
In all three elections, the Democrats set up more field operations than the Republicans, in some cases significantly more, with the Obama campaign being particularly bold in throwing resources behind its field offices during the last two elections. The strategy wasn't just about numbers of offices, however. The real key was operatives getting out in the communities and face-to-face contacts. Phone banks were de-emphasized.
"It was a historic moment for the 2008 and 2012 elections in the way ground troops were utilized," Chung says.
Turning to the effects of mass advertising on the elections, Chung and Zhang compared the effectiveness of ads prepared by the candidate's own campaign with ads supporting the candidate prepared by outside political groups, including political action committees (PACs).
“"I need to see a good ad about Obama to influence my preference for Obama.”
A candidate's own campaign ads were found to be more effective among nonpartisan voters, who had not developed a strong affinity for either party.
It was also learned that TV advertising may not be a deterministic factor when one party has a big advantage, but it was found to be critical for success in a close race like the 2004 election when Republican George W. Bush emerged victorious. Chung says that had Democratic candidate John Kerry received more advertising from outside political groups, the election would have ended up in a 269-269 electoral tie.
"Advertising can shift the needle when it's a close election," he says. "In certain elections, one party has the advantage. If that advantage is very small, because of economic reasons or other factors, then advertising has an effect. If the advantage is big, there is a limit to what advertising can do. That's when field operations become crucial."
The researchers also closely examined outside ads by PACs, which have appeared on many radar screens in recent years. The Democrats had more ads by the candidate than the Republicans in each of the three elections. However, in the case of the 2012 campaign, the PACs supporting Republican Mitt Romney (Harvard MBA 1974, JD 1975) were responsible for 46 percent of his ads, far outnumbering PAC ads for Obama, and as a result, 25 percent more pro-Romney ads than pro-Obama ads were aired.
While the impact of PAC ads in general was not nearly as strong, roughly a tenth of the magnitude of field operations, PAC ads had an effect similar to field operations in that they also tended to galvanize partisan voters. Since PAC ads generally attack rivals, rather than promote candidates, the strong negative tone of these ads seemed to fly better with those who already had strong partisan beliefs, rather than fence-sitters.
"Most of the PAC advertising is very negative. So if I'm already pro-Democrat, a negative ad about Romney for instance will reinforce [my beliefs] and influence me to go out and cast a vote for Obama," Chung says. "But for the not-so-partisan voter, the PAC ad has no effect. If I'm more neutral—an independent voter—I would like to see the vision the candidate has, rather than a lot of slashing and bashing of the other candidate. I need to see a good ad about Obama to influence my preference for Obama."
One other important point emerged in the research. Despite the belief by some that voter behavior is predetermined and campaigns may have minimal effects on converting voters, the researchers found strong empirical evidence that campaigns did in fact matter in the three presidential elections. They played an essential role in determining the outcome of an election, particularly in key battleground states where candidates typically concentrate their campaign resources.
"Marketing is very crucial in an election," Chung says.
Lessons For Business Readers
Chung says the paper may be the first to analyze the effects of multiple campaign activities in such a detailed way. The findings might be helpful to candidates deciding how to allocate their limited resources of time and money, particularly if they appear to be weak in gathering votes from one type of voter or another.
"This could give them a basic approach about how to allocate their funds," Chung says.
Corporate executives can learn lessons about promoting products from this research as well.
"If you think of a candidate as a product—either you buy Obama or you buy Romney—then the logic would hold the same for a company. If you want your loyal customers to buy more, then personal selling is key. If you want your neutral customers to become customers, mass advertising is the way to go. If you have a limited sales force, then you could possibly put out some negative ads against competitors, as PACs do."