One Last Thing to Win: Let’s Go Relational!

Connecting to your own personal network to urge them to vote could tip the 2024 election where it matters most. Read on to learn how.

By all accounts, the presidential election of 2024 is going to be decided by who wins seven states: Arizona, Georgia, Michigan, Nevada, North Carolina, Pennsylvania and Wisconsin. Their 93 electoral votes will determine the outcome. There are also some very important Congressional contests that will decide which party controls each chamber, don’t forget. But at the highest level, the fight for America’s future is now in the hands of voters in those seven states. If you want to make a difference in these final 21 days, start by focusing there.

Just about everyone in the Democratic ecosystem is now in get-out-the-vote mode. Last week, for example, if you lived in Democrat-rich Washington, DC, the top three things the Harris campaign emailed you to go do were: take a bus to Pennsylvania to knock on doors, take a bus to North Carolina to knock on doors, or take a bus (a ten-hour ride!) to Georgia to knock on doors.

Door-knocking is good. So is phone-banking, which you can do from anywhere. If you have the time and inclination and want to help the Harris-Walz campaign or any of the statewide coordinated campaigns, by all means go sign up for some shifts. You can find lots of them easily via Mobilize.us. That said, there’s one additional thing you should do that is arguably as valuable, if not more so: Take your personal contact list, compare it to the national voter file, and find out which of your actual friends, family, co-workers and past acquaintances live in swing states and districts where a call or text from you could be hugely influential.

In a moment, I’ll explain how you can do that easily. First, let me explain why this tactic, which is generally referred to as “relational organizing,” is so potent. Unfortunately, until now, neither the Harris campaign nor the national party committees have put much emphasis on this approach. The reasons for that are a bit frustrating, as I’ll also explain. But it’s still not too late for us to adopt it; it isn’t hard to do; and it can help turn out a few more key voters in an election where that may make all the difference. Indeed, if we grow our relational network large enough, it will help tilt key states. As Sri Kulkarni, a former congressional candidate who is now working on the Democratic Congressional Campaign Committee’s battleground states relational organizing program said to me last night, “If 20,000 people across Michigan, Pennsylvania and Wisconsin spent 15 minutes matching we would be home free.”

Why Talking to Strangers Isn’t Enough

For decades, nearly all Democratic field work has been structured around getting strangers to talk to strangers. That’s what you experience when you go canvassing or phone-banking. And depending on how well these field campaigns are structured and run, they can increase voter turnout by about one or two percent. That said, researchers have also found that these tactics can also backfire and depress turnout or have no discernible effect, and the most recent meta-analysis by the Analyst Institute found, on average, that the positive effects of these tactics, compared to things like warm texting, social pressure mail, or relational organizing, were negligible. (If you want one independent field program that is consistently above par, check out Working America.)

What’s confounding about this approach to activating voters is that everything we know about how people behave and make up their minds is that we are most influential to and influenced by the people we know. And today, thanks to the ubiquity of social media and smart phones, all of us have in our pockets a simple way to access our own personal social universe. But when you go canvassing or phone-banking, no one asks you if you know anybody in the district you are trying to help move. That’s because the core technologies that Democrats have relied upon were built before social networking was so widespread. Worse, because the key voter platform NGP VAN was given a de facto monopoly to be Democrats’ portal to the national voter file, it faced no pressure to innovate even as newer tools and approaches started to proliferate. As I reported back in April, the DNC and Association of State Democratic Parties only ended VAN’s monopoly a year ago, and in that time VAN has just begun opening up its application programming interface to work with newer tools.

As one veteran Democratic political technologist put it to me as I prepared this post, “There is no reason why me sitting here as an organizer in X County, sending around turfs [lists of doors to knock] couldn't coexist with ‘Hey, just turn on this list, and there's a set of people around you that I need to prioritize, and I bet you know a bunch of them, and you talk to them, and we'll all be better. There's no reason why those two types of voter contact behaviors shouldn't be integrated from a technical or programmatic perspective.” Why not?  “The answer is, VAN sucks.”

To put that in plainer English--while data about who we are personally connected to is inherently of great value, the entire Democratic campaign data machine has never been set up to grow that information set and use it in all its critical races. That means when you go door-knocking in a swing state, you may end up talking to my cousin; and when I go door-knocking, I may talk to your brother-in-law; and the fact that each of these target voters could have been reached by someone who actually knows them personally is not integrated into any of these campaigns.

The first time campaigners demonstrated the real potential of relational organizing at scale was the 2020 Senate run-off election in Georgia. In just one month, the Jon Ossoff campaign built and mobilized a network of more than 160,000 voters-- more than half of which were under 40 and 25% of whom didn’t vote in the 2020 general election--by using relational matching and outreach; a later analysis found that the effort increased turnout among those voters by 3.8%. And that worked not only because people were identifying their own contacts; the campaign also ran a sophisticated effort to support those volunteers involving 3,000 paid “mobilizers” who encouraged them to “chase” down their contacts to find as many votes as possible.

Sri Kulkarni, who I quoted above, is a former Democratic congressional candidate who essentially stumbled into relational organizing when he first ran in a very red district around Houston in 2018. Despite having no national support, his relation-centered campaign came within five points of an upset. Two years later, having gotten on the national radar of House Democrats, he ran a much more traditional-style campaign that emphasized fundraising and paid media, and actually did a tad worse (his was a very tough district for any Democrat). He also worked on the Ossoff relational push in 2020, and then built a larger effort in Texas in 2022. (Unpublished data from that Texas project found that relational outreach outperformed traditional door-knocking by more than 20 points.) As of a few weeks ago, he was hired to run the Democratic Congressional Campaign Committee’s relational organizing program. And as a strong advocate for the tactic, he doesn’t hold back his frustration with how little it is being used this year.

“Most campaigns that I talk to [about relational organizing] say, well, it doesn't really scale up the way that advertising or phone banking does,” he told me a few days ago. “The problem is that they're looking at it very incorrectly. What we did in Georgia was we pushed people not to just think of people off the top of their head, but just try to match as many contacts as they have, and we were getting their networks up to 50 or more [per participant] in Georgia in 2020.” What that means is the odds of finding previously unidentified potential voters goes way up. In 2022, he said, “I did it again in Texas, and Texas is a much larger state, so you have more people to match if you know folks in Dallas or Houston or Austin. And we got people's network sizes to an average of over 100.” In his own case, he built a network of 3000 people, who were registered voters in Texas that he had on his phone. “And out of those, I found 400 people who had missed the previous election….And most of those were not actually my friends, directly, they were my friend's husband or mother-in-law or son or something like that.”

The good news is that there are newer tools that do make network-based voter contact easy-- most notably Reach, Rally and SwipeBlue. Reach was first built by technologists supporting Alexandria Ocasio-Cortez’s 2018 upstart congressional campaign, to address the fact that many of her supporters were not in the Democratic voter file and they needed a way for canvassers who might be meeting potential voters on subway platforms or at food markets to add people to their list. (No, VAN does not make that easy!) Now Reach is a more robust app that lets any user match their own contact list against the Democratic voter file and then build a personal network of people who they can reach out to directly.

According to Kulkarni, about 50,000 people have used Reach to join “The Democrats” on the app, which is the DNC’s account and which is now being used by the Harris-Walz campaign. (You can find the Harris-Walz Reach link from the bottom of this Go.KamalaHarris.com page) So far, most have not matched their personal network to voters, instead mainly using it to share timely social media content with their networks. There were only about 500 people signed up for the DCCC’s “Take Back the House” campaign on Reach when Kulkarni came on board to goose its efforts — now they have doubled their users and reached eight times as many voters. If more people get trained, there are potentially tens of thousands of extra votes to be had.

There’s one big hurdle in the way of getting more people to match their contact lists: people fear losing control of their personal information. Many of us are now allergic to any request that we “share” our contacts, having been taught that Big Tech companies are just selling our information to advertisers. People also hate how often they are spammed by Democratic campaigns and fundraisers—once you donate to one it seems everyone sooner or later has your phone number and email address. I myself have pooh-poohed some “relational” programs because I saw them as just a way for campaigns to get ahold of and exploit people’s personal networks. But Reach and SwipeBlue are different: when you create an account you are not giving your contact list away; you are being given matching information that you can use for your own outreach.

Kulkarni has a vision for the next three weeks. If enough people get on Reach and join the DCCC’s “Take Back the House” campaign, on November 1 it may be able to give every congressional campaign a list of missing voters in their district who have not yet voted and who are “in network” and reachable by volunteers who know them. That requires more than simply signing onto Reach; participants are being encouraged to “friendbank” and get more of their contacts in swing states to also get on the app. Every night, he and his team are running “Friendbanking” workshops online that will get you trained in how to use Reach for this specific purpose. Sign up now. Or go check out Rally, which is a paid relational program being run by Relentless, a consulting shop made up of many of the people who built the Ossoff relational effort.)

SwipeBlue is a somewhat simpler app that also invites users to match their contacts against the voter file that was built by Democrats.com, an independent group launched years ago by Bob Fertik. “We built SwipeBlue for truly simple relational texting,” he told me. “It's as easy as Tinder.” He says that typically, a user will match around 50% of their contacts to the voter file. The app shows them which are Democrats and where they live. “That lets us create campaigns at all levels, including Kamala Harris battleground states, Senate battlegrounds, and abortion referenda,” he said, adding, “We also have statewide campaigns and US House campaigns for NY and CA, which will be key for flipping the House.” As of now, close to 38,000 people have installed SwipeBlue. “Activists can easily use SwipeBlue on their own,” Fertik says, “but we are trying to make it fun by hosting VoteRaisers where we text our friends together. Last week, Jane Fonda, Lily Tomlin, Kevin Bacon and other celebrities led a “Six Degrees of Kevin Bacon” vote-raiser texting friends in the battleground states.

To be clear, what we are talking about here is still not the kind of humanized community organizing that we desperately need to rebuild and sustain social solidarity in the face of rising fascism. The Harris-Walz campaign, DNC, DCCC and just about all the congressional campaigns people are trying to help are not going to spend a penny past November 5 to sustain any of the networks being built now. They’re just building big sandcastles that will get washed away. Even relational organizing as its being practiced by Kulkarni and Fertik is still just an augmentation of the atomized marketing culture that permeates Democratic party work, which my friend Marshall Ganz so eloquently critiques. Though with Reach and SwipeBlue, once you build out your own network of high-value contacts, it’s still yours. You can’t say that about the people you may talk to as you door-knock or phone-bank.

Why haven’t you heard more about relational organizing and its potential in 2024?

There are a couple of reasons. Almost no one in the political media covers organizing, and if you do see a story about field organizing, it’s invariably just a color piece with mere whiffs of hard data. It’s a lot easier and cheaper to report on the latest poll or campaign ad.

Also, let’s face it, the Harris campaign is the Biden campaign organization with a different candidate. Remember how few people Biden had on staff as late as last December? His team was never imagining that it needed to create a structure that could absorb and deploy hundreds of thousands of volunteers, and its struggling now to keep up. People who are currently doing a huge amount of gap-filling with groups like "Xes for Harris" (see ForHarris.org) have told me they can't even figure out who is in charge of coordinating volunteers for the Harris campaign. 

Most campaigns don't value field organizing; it's considered low status if you are trying to advance in your career as a professional political operative, and you don't get a 15% cut for spending campaign funds on field compared to paid media. Yes, the DNC is spending tens of millions on canvassing, but all of that is flowing to intermediary vendors who hire lots of young kids at poverty wages. Hiring paid canvassers via some of the wholesalers that now specialize in that work doesn't produce quality canvassing. It does, however, check a box on someone's metric.

Efforts to use new, proven techniques that are more productive (like relational organizing and deep canvassing) have been only modestly embraced. That's partially because they require some training (which can be relatively expensive), though there are free trainings available--the Harris campaign so far isn't pointing people to them.

My bottom line conclusion: if you can go door-knock in a swing state, by all means go. Or, take the money you would have spent on the trip and give it to an organization in the state (the Movement Voter Project will give you a good list). But IN ADDITION, there's still time for everyone to add their social graph to the relational programs being run by the Harris campaign and the DCCC on Reach and Democrats.com’s SwipeBlue app.

Related:

—MoveOn is urging its members to use vote-tripling to help find and activate missing voters as the race comes into the final weeks. This is relational organizing at its most stripped-down. Volunteers and paid staff ask voters, either people they connect with at the door or on the street or literally after they’ve left their polling place, to text three friends to urge them to vote. It helps. But it’s really very transactional. It takes very little training to do and it also builds nothing lasting.

—According to the CNN, the Trump campaign is reportedly relying heavily on a "10 by 10" strategy of each supporter tapping ten more people they know. That's relational at its core. I have no idea if they have the ability to pull this off or if it will have organic traction. Generally speaking this isn’t something people do very well without help and guidance. In some ways, this hyper-close election is coming down to a test of two very different ways of getting out the vote. One is the professional data-driven approach; the other is the appeal to raw emotion and social connections.

—The Guardian’s Hugo Lowell reports that slow internet connections in rural areas are hampering the efforts of both the Trump campaign and Elon Musk’s America PAC to use a campaign management tool called Campaign Sidekick to track the efforts of their canvassers. Apparently the app forces door-knockers who have slow bandwith to use offline walkbooks that have no geo-tracking features. This means headquarters has little way to know if they’re actually knocking on doors or cheating by tossing campaign materials while driving past them. Trump’s co-campaign chief Chris LaCivita denies that they’re having problems. The campaign isn’t using tools like i360 Walk, which is made by the Koch network, because of past animosity between Trump and the Koch brothers. Can we thank the campaign gods for little things?

One thing if you’re thinking ahead

—The Democracy Fund has launched Election Day to Every Day, a new effort to ensure that pro-democracy organizations don’t experience a sudden dropoff in funding after November. DF president Joe Goldman explains more here.

The Last Word

The Connector just topped 7,000 subscribers, triple the number I had when I started this thing in the fall of 2020. Thanks especially to the 309 of you who are paid subscribers—you make it possible for me to keep doing this work and to keep all the contents of this newsletter free to all readers. Please share this issue widely, if you can; the more we can get people friendbanking, the better the odds on November 5.