Planting Liberally

Recently the Obama campaign quietly released the Neighbor to Neighbor tool, an innovative approach to field work which releases volunteers from the need to go to a campaign office in order to reach potential voters and volunteers. This tool has been kicking around Democratic circles for a while - first in the Lamont campaign's postcard tool, then in MoveOn's phonbebanking system for the 2006 general election, as well as Deval Patrick's DIY canvassing effort in the run-up to his landslide victory in Massachusetts.

The basic idea is simple. If you want to volunteer for Obama, just go to my.barackobama.com, and either sign up or register for an account. Once you're logged in, you'll see a list of "Neighbor to Neighbor" campaigns on the left hand side of your screen; click one of them, and the website will take you through the necessary next steps. At the end of the day, you get a list of people who the campaign needs to contact - either prospective volunteers who you could bring on board to increase capacity, or voters who you could convince to vote for Obama. You also get a script to use when you're making calls. When you're done with the calls, you record the results of each call (Was the person home? Will he or she volunteer / vote for Obama? etc.). The campaign has a good video explaining the process, too:

I'm very pleased to see this system come on-line. It's an excellent way to empower volunteers and to radically ramp up the campaigns potential for volunteer activity. If you're not signed up at my.barackobama.com, head on over there and register now. Then schedule some time to make Neighbor to Neighbor calls in the next week.

Despite my enthusiasm for this system, I think there are a few things the campaign could do to improve upon it, if there's time:

1. Publicize it better. I've seen almost no mention of Neighbor to Neighbor anywhere, except for a passing reference on a blog post (and I can't remember where that was.) I had some idea that this would be coming online eventually, since I remember hearing about it during Patrick's campaign, and I can't imagine that the Patrick campaign had a single tactical or strategic innovation that wasn't shared with Obama. But a Google search for "obama neighbor to neighbor" turns up a good post by Jack and Jill, an embed of the video clip I posted above - and very little from the Obama campaign itself. Despite having attended a couple of volunteer organizing meetings in my neighborhood over the last week, I've heard nothing about Neighbor to Neighbor. Why not publicize this great new system on the email list, or at a bare minimum make volunteer organizers aware of it?
2. Make it Facebook-savvy. I've installed the Obama Facebook application, I've registered on my.barackobama.com. Why did I have to do the two things separately? Much more importantly, how come the one doesn't seem to know anything about the other? Why isn't My.BarackObama.com trolling through my Facebook friend list and asking me to invite those of my friends which are also on its list to register to vote, volunteer, or otherwise get engaged? Why can't I find out which of my Facebook friends are listed as undecided, so I can chat with them about the election, or possibly clear up a gap in the campaign's records? As far as I can tell, the Facebook application is just a content delivery mechanism, which seems like a serious underestimation of Facebook's organizing capabilities.
3. Better matching capabilities. What I've heard from professional tele-fundraisers is that the best people to staff the phones on a campaign are either those who are naturally good at telemarketing, or those who are demographically similar to the target population. It seems obvious enough, but that kind of smart matching rarely happens on political campaigns. Currently, the Neighbor to Neighbor program matches me with other people in my geographic area (since I asked to speak to prospective volunteers, anyway). That's a reasonable way to approximate demographic matchup, but there are plenty of people who live near me but aren't at all like me. Potentially, a web developer in Ohio or a Jewish grandmother in Florida would be a much better person for me to talk with than a lawyer down the street. Yet the system doesn't ask me anything about my occupation, religion, racial identification, or other demographic indicators, and I can almost guarantee that on the other side, there's no cross-referencing of voter registration records to commercial databases that could reveal similar information about voters.
4. Open up the data. From what I can tell, there's no way to get this data and write a program to do something interesting with it. That's significant, as both of my last two points could be addressed by a sufficiently energetic team of developers, without the supervision of the campaign, writing data-mining or Facebook-mashing applications to make the Obama campaign's database come alive. The point is, these two ideas might be the tip of the iceberg, and there could be other, smarter applications waiting to be unleashed. This point is all the more significant because, I'd wager, data-miners and web developers are probably emphatically pro-Obama. The weight of technological innovation is squarely in Obama's camp this year, and the campaign should press that advantage to the hilt. I recognize there are important privacy concerns regarding this data, but there must be some way to properly license or protect the data while allowing outside developers to innovate on top of it.

At this point, Obama's exceptionally strong ground game could easily be the difference between victory and defeat. Neighbor to Neighbor could be a game-changing application that blows open the potential for volunteer engagement in the campaign. It's a wonderful tool, but it needs a bit of tweaking at the margins to really make it shine. I'd love to hear from others - have you used Neighbor to Neighbor? If so, what are your thoughts or critiques? Any thoughts about Obama's ground game from an in-the-trenches perspective?