Adventures in Microtargeting

In this much-discussed recent article, Sasha Issenberg argues that Democrats are much better than Republicans at “microtargeting”: the art and science of tailoring messages to the proclivities of individual voters and contacting them directly by phone, mail, or sending a volunteer. Issenberg’s account is interesting and entertaining. And it’s certainly possible that the Democrats are better at this than the GOP. So far, however, it doesn’t seem to have had much effect. The Democratic victory in 2008 was roughly in line with what might be expected, given the political and economic situation. Microtargeting didn’t keep the Democrats from getting clobbered in 2010, when the economy and some other factors worked against them. If Obama wins by a narrow margin this year, that result, too, would be in line with historical expectations. So far, at least, it seems like microtargeting is no more than a minor Democratic advantage.

Living in the swing state of Virginia these past few months, my wife and I have had more exposure than we might like to both Democratic and GOP microtargeting. Neither strikes us as especially brilliant, though the GOP seems worse. Perhaps because we are on various Federalist Society and libertarian mailing lists, we have been deluged with mail from the Romney campaign and allied organizations. Some of it is at least reasonably calculated to appeal to us (e.g. – promising to cut federal spending). On the other hand, a lot of it consists of attacks on Obama for failing to impose trade sanctions on China and promising to punish the Chinese if Romney gets elected. It’s hard to hit on a message more likely to alienate two pro-free trade libertarians. If we believed that Romney really plans to start a trade war with China, that would be a big strike against him in our minds. My guess, however, is that his promises will likely go the way of the very similar protectionist prattle that Obama put out in 2008.

The Romney campaign and its allies have also deluged us with robo-calls. Since we generally hang up on them, I can’t really assess their content. However, the very fact of making dozens of robo-calls is likely to annoy voters far more than it attracts them – even if the content were truly brilliant, which is rarely the case.

We have also been visited by at least one Romney volunteer, with whom I had an interesting conversation. When I saw him come to the door, I immediately said that I might end up voting for Romney as the lesser of two evils, but I have no real enthusiasm for him. “Why not,” he asked? I explained it was because of Romneycare and a litany of other flaws in Romney’s record. To my great surprise, the volunteer readily admitted that he wasn’t enthusiastic about Romney either, and for many of the same reasons. In general, it does not make a good impression if a campaign’s volunteers admit they don’t really like the candidate they’re supposed to be promoting. In this case, however, the lukewarm volunteer made a far better impression on me than a gung-ho enthusiast would have. Given his record, no one could convince me to actually like Romney. I might, however, be persuaded that he’s a lesser evil. If the Romney campaign deliberately sent me this person, it was a (very minor) stroke of genius. But I suspect it’s far more likely that they simply had no idea that that this volunteer didn’t really care for the candidate.

Still, the GOP did correctly identify us as likely supporters, and correctly focused on economic issues in its messages to us. My wife has always been convinced that Romney is the lesser of the two realistically electable evils. I am far less certain, but am just barely inclined in the same direction. Yet if the Romney campaign wanted to shore up my wavering support, they didn’t choose the right strategy for doing so.

We have also gotten some Obama mailings and flyers, along with a few robo-calls. All of them seem pretty generic, and many emphasize issues where we and Obama are clearly at odds. To the extent they had any impact on us at all, it was to remind us of the many things we don’t like about the Obama administration. We also got a visit from an Obama volunteer. When my wife told him she was planning to vote for Romney (I wasn’t home at the time), he immediately said good bye and left.

I suspect that the Obama campaign did not actually target us as carefully as the Romney people did – perhaps because we either didn’t show up in their databases, or showed up as people they couldn’t hope to win over. As a result, they missed the fact that I was actually a potential swing voter, even if my wife wasn’t. Still, they at least wasted fewer resources on us than the Romney people did, and said fewer things that were likely to actively annoy us. And I certainly appreciate their relative restraint on the robo-call front.

It would be easy to make fun of the campaigns and their often inept outreach to us. But it’s dangerous to generalize too much from a single case. The real lesson, if there is one, is not that the campaigns are stupid but that the effective micro-targeting is very difficult when you’re trying to process and act on information about hundreds of thousands of voters in a relatively short time. Had they taken the time to do their homework on us, both campaigns could have made much more effective appeals. Unlike with most voters, in our case there’s actually a lot of publicly available information about our political views. But the campaigns simply don’t have the resources to carefully study every voter, even in a closely contested swing state. For that reason, micro-targeting is likely to remain an inexact science at best. And I remain skeptical of claims that it is going to be a major decisive factor in winning national elections.

UPDATE: I think microtargeting could potentially be more effective in small-scale, local campaigns, where there are far fewer voters to study. On the other hand, few if any local campaigns have the microtargeting resources available to national campaigns.