How to Be Leaner and Meaner with Targeting and Data

The New York Times magazine recently ran a 10 page feature giving an inside look at how the analytics team at the Obama campaigns of 2008 and 2012 (alternatively described as "huge R & D projects") turned the TV marketing world on its head by being extremely savvy, and precisely targeted, with data. There is much to be learned from in here, from the angle of film distribution. Firstly, in detailing the corporate nature of some of the Obama team's lives after the campaign, the article makes clear targeting seems to be an example of one kind of tactic that is relevant in a for-profit venture as much as a non-profit political campaign (although maybe we should not be so quick to lump "non-profit" and "campaign" together all the time -- as the author puts it in this article, campaigns are more like "start-ups aimed at a one day sale," inadvertently also highlighting another similarity they have with a film's distribution). Though in former posts we've contemplated the state of the independent film world as caught in a binary between a for-profit venture and a donation-based arts-patronage endeavor, more recent posts about the Scottish Documentary Institute and NationBuilder demonstrate that some organizations are giving all entry points a seat at the table, at once. Targeting works from either angle.

The thrust of the article basically conjures the campaign's process in finding a cheaper and more effective way to use data to identify who they should be targeting with their persuasion tactics. Through cross-referencing data sets from voter contact, Facebook, party-voter lists, and a new TV tracking firm, the campaign ended up spending 35% less per broadcast than the Romney campaign, but got 40,000 more spots on the air for $90 million less. (Truly astounding). The ultimate irony, from a film business lens, is that that tracking firm they used is Rentrak -- which, while also being a competitor for Nielsen to inform TV viewing practices, has been the film industry's standard paid-for service to measure how films perform in each of their theatrical venues and markets. In other words, the campaign utilized one of the film industry's tools, but more effectively than the film industry itself, the big studios of which likely (and this is conjecture) use what the article classifies as the more antiquated, less reliable, and certainly less precise Nielsen system to determine where to throw their advertising dollars (combined with "hunches and deductions"). Nielsen is in 22,000 homes; Rentrak is in 8 million. Nielsen breaks down audiences into big chunks like "18 to 49 year-old male"; it did not provide a proper data set to cross-reference the 15 million potential Obama voters the campaign had identified through Facebook. Nielsen is what tells advertisers and studios to broadcast an ad aimed at a huge swath of the U.S. during primetime; the campaign's Rentrak-based data told them they could do more for less by targeting "Judge Joe Brown" and "The Insider" viewers in early afternoon and 1 A.M., respectively.

This is where the clunky, old school, risk-averse practices of industry entities like studios put them at a potential disadvantage to independent filmmakers - if, and only if, independent filmmakers can be nimble with data like the campaign was. Of course the campaign was a national entity with massively widespread recognition for the "movie" they were selling, and it had plenty of money for its operation, but targeting is desirable because it uses less resources to get more results. This is the bedrock of a grassroots philosophy. Admittedly, this article centralizes one specific format for advertising: television, which requires more money than most grassroots filmmakers will have in their budgets in order to use it to advertise. But both the origin story of where they got the data that led to the TV ads and how they thought about their TV ads are relevant.

For one, Dan Wagner, the analytics guru featured in the article, notes the sea change in the political data world from the Bush years, where they would divine how a person would likely vote from their car and sports preferences. "Why engage in such divination when you have the time and money to just call voters and ask them about their leanings directly?" he asked, describing what they went on to do. Both Obama campaigns oriented their volunteer force around giant voter outreach operations, which supplied much of the data that people like Wagner had to work with -- specifically, who was undecided or should otherwise be targeted for persuasion. Another caveat is deserved to say that Wagner does attribute this to "time and money," two things that a grassroots filmmaker likely won't have much of, but it is still relatively cheap to pick up the phone and ask someone about their preferences. Could this be applied to a film's release? A political campaign can start with a party voter list; how would a film campaign know what numbers to call? The answer, as it was for the Obama staff, may be in cross-referencing. See how far one gets through social media -- not everyone who likes a film on Facebook is going to turn out for it. They, and their friends who may have the same film-going interests as them, provide a place to start. Also, in doing grassroots outreach to organizations based on an "issue" or subject matter that might be of interest to them, filmmakers are already sort of doing "targeted" cold-calling. It's just a matter of prioritizing phone conversations like these, as well as mining the aforementioned social media outreach for potential audience members. Also, though the idea of going door-to-door to gauge or spread interest in an independent film that is premiering locally seems like madness, if other data sets generated a targeted map to go from one house of a likely independent film watcher to another, it might seem like more than just a drop in a bucket.

The Obama campaign spent 50% of their budget on advertising. That might at first read as a stark rebuttal of the importance of their field operation. But for grassroots filmmakers, outreach like this is the equivalent of advertising; there is no reason not to prioritize it just as much as the campaign did. In fact, if filmmakers save enough money by doing outreach, they may even have enough left over to actually advertise on TV themselves! If research yields that one spot on a certain show at a certain time would give enough bang for the buck, why not?

Bringing the parallel worlds full circle, the folks at Wagner's new venture, Civis Analytics, seem perfectly set up to be used by filmmakers: they aim to "use analytics to help nonprofit and for-profit companies reach out to segments of the population they were struggling to connect with."