Door-knocking is the best way to earn votes, but for all their vaunted tech savvy, Democrats’ core tools and voter data are a mess
Sunday I spent the afternoon walking the hills of ex-urban Woodbury, NY, about an hour north of New York City, knocking on doors with about 30 other volunteers from my local Indivisible group, NYCD16-Indivisible, along with a smattering of other grassroots activists from Westchester county. We were there to talk to registered Democrats about several candidates, starting with Tim Ryan, a veteran who is running in the newly redrawn 18th district, along with several of his down-ballot colleagues including James Skoufis, an energetic two-term state senator.
It was a gorgeous day to be out and about, vibrant with the yellow-orange-red hues of almost-peak autumn in the Northeast. And it was even better to be door-knocking with friends, people I’ve been working side-by-side with since late 2016. (Canvassing together is much better than showing up at a random place to do it solo.) All told, our group, which broke into pairs and got lists of addresses all in the surrounding verdant hills, hit more than 650 doors in total and tallied 168 conversations. According to the field staffer overseeing the afternoon, we identified 130 supporters for Ryan and the other candidates, which is pretty good considering that our list was focused on infrequent Democratic voters, not the people who reliably have voted prior elections.
Sadly, MiniVan, the app we were using to walk our designated turfs and tally our interactions, is still just as frustrating to use as ever. There’s still no built-in way to add new names to the file, which is madness. Nor has NGP-VAN, MiniVan’s maker, or the Democratic Congressional Campaign Committee, the sponsor of this particular canvas, done much to fix ongoing errors in the Democratic voter file. Of the 30 doors where I had a conversation, at six I was told that the person I was asking for had moved, and in three cases “ten years ago.” Later, when a young Hispanic dad playing with his kids saw me pass in the street, he called out to me to say that “my whole house is voting your way.” I realized he was one of the people whose door I had knocked earlier only to be told that my target had moved away. In some alternative universe run by a state or national party interested in growing its support base, MiniVan would make it super-easy to add new names and correct bad info. But unless a campaign opts to add the “Street Team” package, by default that feature is off.
On Google Play, Minivan has a lot of mixed reviews and an overall rating of just 3.2 out of 5. On the App Store, it does only a tiny bit better with a 3.3. Why isn’t this core piece of Democratic political technology better? And why is the voter file still a mess? Two answers. First, monopoly. Second, that’s not my problem.
A source of mine who used to work for NGP-VAN but left after the company was acquired last year by the Apax investment group, becoming part of a software conglomerate called Bonterra, tells me we shouldn’t expect to see significant improvement in Minivan because none of the big players in the field of political or nonprofit engagement want to pay much for that. Digital fundraising will continue to draw the lion’s share of investment because frankly, it makes more than it costs. Even worse, the decline in support for Minivan began before NGP-VAN was acquired, as part of an across-the-board attrition in engineering headcount that made the company look more appetizing to buyers. As of February 2020 VAN‘s organizing feature software team was disbanded, leaving only one team of engineers working on Minivan. Will this get better anytime soon? Not when national Democrats and the Association of State Democratic Parties continue NGP-VAN’s privileged position as the monopoly provider of this software for voter engagement, and when the private equity overlords at Bonterra see expanding their market share in the nonprofit world as a higher priority than continuous improvement of core political tech.
As for fixing problems with the voter file, I’ve seen enough campaigns now to say that the reason nothing gets better is that no one owns this problem. Individual campaigns just need enough votes to win their campaigns; making sure new names get inputted properly just isn’t that important to them when they are racing to election day. Most campaigns don’t worry that the information canvassers collect isn’t perfect as is — did I really ask every person who I talked to if they were indeed voting for Assemblyman Chris Eachus or Bernie Silver for sheriff, let alone if their support was strong or just likely? No, I didn’t nor did any campaign staffer make any effort to make sure our canvassers were following their scripts that scrupulously. The numbers we reported will get duly reported to those campaigns, but they’re as solid as warm butter.
In theory state parties would do the work of maintaining, expanding and improving the quality of their voter files but again, I’ve seen no evidence that this is a high priority for them. Civitech, a political software start-up led by Jeremy Smith, has been investing heavily in scouring other sources for the names and addresses of people to be added. He tells me while the Democratic voter file contains about 195 million records, about eight to ten percent of those are wrong. On top of that, there are many people who get skipped over by nearly all campaigns because they aren’t in the file in the first place. Smith says his team has found about 18.6 million names and addresses that could be added, out of the roughly 64 million adult Americans who are current unregistered but otherwise eligible to vote.
As a fledgling competitor to NGP-VAN, Civitech’s ability to help campaigns reach these otherwise neglected potential voters ought to be a big selling point. But that assumes that Democratic politicians actually want to grow the electorate they aim to represent. And that gets to the last dirty secret about too many incumbents: they’re more comfortable getting the votes of the voters they already know (and have built a reputation with) than with earning the votes of new voters. New people voting often make new demands that incumbents don’t know how or don’t want to handle. And so even if a bigger electorate would be good for democracy, it’s often the last thing incumbent politicians and their allies in state parties actually want to see. Someday we’ll fix these problems, but that day will only come when, instead of trusting the consultants in the political industrial complex now in charge, grassroots Democratic activists take control of their own destiny.