We Knocked Too Many Doors

It started last January. In my work as a campaign strategist, I had a dozen conversations with the leaders of membership-based organizations who do a combination of year-round issue-based organizing and independent get-out-the-vote work in election seasons. Almost all of these groups have emerged in the last ten years, as the consensus within the ​“organizing left” on electoral participation shifted from ​“maybe we’ll hold a nonpartisan candidate forum” to ​“we’ll (ideally) recruit our own candidates and devote significant resources to getting them elected.” 

In every instance, I was discussing the feasibility of launching issue-based campaigns that would eventually set up the election as a referendum on affordability. The idea was to press government officials for plans to bring down the cost of living and then contrast the lack of competing (Republican) proposals. They would have called on governors to make commitments to stand up to investor-owned utility companies — which, in most states, are allowed to charge us forany costs they incur building dirty energy plants — or asked county commissions to use federal funds to lower the costs of operating public schools, which could in turn lower the amount local taxpayers are asked to pay. To be sure, none of these campaigns were a slam dunk – and they would have required significant education of both members and Democratic elected officials. But that wasn’t the reason these organizers were saying ​“no.”

“I can’t think about that until after the general.” 

“We would need the campaign to wrap up in the spring to have enough time to shift to electoral.” 

“The table wants us to knock at least 400,000 doors in the fall, so I think we would have to shift to field by mid-July.” 

“Our organizing team would have a hard time pivoting from an issue fight to messaging around the election.”

I agree with the self-assessments that these groups would not have been able to do both get-out-the-vote work and an issue-based campaign at scale, simultaneously. And before November 6, I wasn’t sure that there was a clear-cut cost to prioritizing door-knocks over organizing committees in 2024. 

But it’s clear now that, absent a transformational Democratic candidate who could make a credible argument that they would shake up the economic status quo, there was a significant electoral cost to that choice. Paid outreach simply wasn’t able to make up for weak federal candidates. It narrowed the turnout gap some in swing states but not enough, while groups that had setup their get-out-the-vote efforts as the culmination of ongoing local issue campaigns had more success electing or defending aligned lawmakers.

Before you read this as another left-bashing-left thinkpiece, this is also honest self-criticism — I co-founded a North Carolina membership-based organization, Siembra NC, that just invested a tremendous amount of its capacity to reach 125,000 knocks. In informal conversations with leadership throughout the year I agreed with their assessment about the need to make that investment. Staff and members are now reflecting on whether the investment was worth it relative to the other opportunities they turned down — which might have also developed their leadership skills, recruited (low-propensity voter) volunteers and helped build long-term political power. 

There’s a ​“turnout math” involved in deciding how to use a grassroots organization’s resources to both influence and get out votes in a way that complements the voter contact a candidate’s campaign might do. 

Sometimes all potential voters need is a ​“nudge” — simply asking some people if they plan to vote can make it significantly more likely they will. That’s easy for a campaign, with its vast resources, to do. 

But research on voter behavior shows that in communities less likely to turn out to vote, motivated volunteers from the community are more effective at raising voter turnout. Cultivating those volunteers is something independent organizations are uniquely situated to do. When we started Siembra NC eight years ago, our members – who largely did not participate in political activity before we recruited them – would say ​“yes” to opportunities to turn out voters for elections after being involved in fights to stop sheriffs from collaborating with ICE, win protections for mobile home tenants and recover stolen wages.

Similarly, New Virginia Majority, a civic engagement organization for whom I managed a get-out-the-vote doorknocking program in 2009, recruits far more people — Latino tenants, South Asian cab drivers, Black college students — to develop their skills as member-volunteers through issue-based campaigns than through broader invitations to get involved in presidential get-out-the-vote activities. In a year like 2024, when a huge number of progressives were discouraged from voting by the Biden Administration’s policies, the political education conversations and debates they were exposed to as NVM members motivated many to continue showing up for volunteer canvass shifts. The local campaigns — around rights restoration for returning citizens, affordable housing funding, or shutting down environmental hazards, for example — often shape the story about local and statewide elections, requiring candidates to respond to urgent concerns and pulling in active supporters who might otherwise be ​“on the fence” about voting. In some years the losses outweigh the victories, but using this model they’ve elected endorsed candidates at all levels of the ballot, including flipping their state legislature, and helping pass transformative legislation. 

Paid canvassing is where the turnout math gets muddier, and where independent political organizations can end up duplicating the work of campaigns. Historically, when those independent operations were dominated by labor, union members did a mix of paid and volunteer canvassing, but with an implicit understanding that they had experienced the power and importance of collective action through campaigns for higher wages and better working conditions. This helped paid member-canvassers have the same persuasive power that we see from motivated volunteers.

Based on my conversations with the dozen organizations I spoke with, paid canvassing became the dominant mode for independent political operations in 2024. Facing national headwinds — when so few progressive activists had any interest in getting out the vote for Biden or Harris — base-building organizations invested their limited capacity in paid canvass programs to drive up participation. New Virginia Majority’s co-director, Jon Liss, estimates that powerbuilding organizations in battleground states averaged 90% paid door knocks in 2024. 

Between $1.4 billion in advertising and those paid doorknocking programs, Harris lost voters in battleground states at half the rate of the country as a whole – a 6% swing from 2020 nationwide but about 3% in the states with the most voter outreach. 

But without more for thousands of paid canvassers to talk about beyond repeating the uninspiring message from the top of the ballot, Harris still lost.

That dynamic wasn’t true everywhere – for some groups, year-round issue-based organizing drove volunteers to get out the vote, and paid canvass programs were an add-on. 

Faith in Minnesota’s organizing committees are arguably the only reason Republicans didn’t win full control of the State House, where they are now tied with the Democratic-Farmer-Labor Party. In the eight closest house races, DFL candidates won by anywhere from 14 votes to 1,236 votes — and in six of those eight, ISAIAH’s Muslim Coalition had live conversations with 900 to 2,700 voters. Some of the candidates outperformed Kamala Harris by 2% or more. In the year leading up to the election, the group’s organizing committees ran campaigns demanding culturally relevant affordable housing, interest-free home ownership loans and youth programing investments. All of the candidates they backed championed one or more of these demands, which surely changed the math for some voters about whether to show up to vote, to say nothing of whether to encourage their friends to join them. The organizing committees also drew out motivated volunteers — a majority of Faith in Minnesota’s 50,288 conversations with voters were made by volunteers, who filled 3,422 shifts. 

A similar dynamic emerged in Los Angeles, where tenant organizers have aggressively worked to build public support for local and statewide policies to bring down the cost of rent. Those organizers used their door-knocking muscle to win funds for homelessness prevention and help a tenants rights attorney unseat a sitting council member whom they saw as pro-landlord. 

But for many organizers I spoke with throughout 2024, especially in the seven swing states, the pressure to commit to sizable ​“door goals” early in the year superseded investment in issue campaigns that might redefine the election itself. 

Here’s why I think investing in volunteer membership recruitment via issues-based campaigns would have been a better investment for most organizations that had to choose between the two.

One-off conversations aren’t as persuasive as we think. I know everyone reading this who spent some time on the doors this election — including me – has examples of conversations where our ten- or twenty-minute chat seemed to turn things around for a particular voter, and surely the results would have been more dramatic without our work. But as writer Micah Sifry points out in this anecdote from the field, deep canvass experiments have revealed, in many cases, those voters were not as committed to their (new) position after our conversation as they stated: 

A conventional canvasser might mark a voter as a ​‘strong supporter’ in MiniVAN after a brief chat at their door. But someone trained in deep canvassing, like my friend Kipchoge Spencer, said he discovered both in Philly and in the nearby suburbs that when he spent more time talking to a voter, someone who started out saying they were ​‘strong’ for Harris oftentimes was actually conflicted or not strongly motivated.

This is a difficult pill to swallow, and to be clear, I spent several dozen hours on the doors in the weeks before the election, and I wouldn’t have changed a thing — there was almost certainly no better way to impact the outcome at that point. 

One-off conversations aren’t as persuasive in general elections. There’s a lot of conflicting research about how effective voter turnout is for different kinds of prospective voters, although many scholars agree nearly every tactic is least effective in a ​“high-information” presidential general election, as opposed to local primary elections, for instance, in which opinions haven’t been as thoroughly shaped by advertising and news coverage. And the kinds of canvassing that most approximate face-to-face organizing are most effective — high-quality relational and deep canvass programs. ​“Relational” just means helping someone you already know make a plan to vote, and ​“deep canvass” is basically having an organizing conversation with a stranger, trying to find out what motivates them on a deeper level before bringing up the election.

But when donors are asking field programs to focus on raw numbers of voters reached, these types of canvassing incredibly difficult to prioritize. Again, Sifry has good anecdotes about why that is, but he implies that Democratic insiders might be the only culprits — in my experience, there’s actually a lot of participation by the donors to and leaders of independent political organizations in shaping the expectations that total knocks or conversations are the ideal standard of measurement. The donors, obviously, have more power in this relationship, but those of us in leadership do have agency to push back.

We’re competing to build trust, not echo talking points ​“live.” Most Democratic field programs treat canvassing like an in-person complement to TV ads — another way to deliver talking points. But those of us who come out of grassroots and labor organizing think about it differently; we’re trying to win prospective voters over to our worldview. Nearly everyone I’ve talked to who spent a significant amount of time canvassing reported frustration reaching Harris-skeptical voters who had their minds made up about Trump’s potential impact on the economy, almost all of whom pointed to social media posts as their source of truth. The most-followed online influencers have built trust, just like Trump has built a brand — with constant repetition, allowing followers to get to know them and enter their worldviews over time. This isn’t a new insight to veterans of hard-fought healthcare and public education union fights, where organizers often have to build supermajority unity between hundreds of workers who have sharp political disagreements. Building trust takes time. I’ve helped coordinate general-election get-out-the-vote programs since 2004, and I’ve never seen one that built trust through one-off conversations in a way that could rival what the MAGA-adjacent social media echo chamber has now achieved. What if field program donors instead asked independent organizations to prioritize recruitment conversations for some percentage of their contact goals? 

Without a clear national storyline to amplify, we need to push local demands. Everyone reading this probably experienced firsthand the impossible task of making the case that Harris would represent a dramatic change to voters worried about the present, not a possible future under Trump. Here in North Carolina, absent local issue campaign demands in most electoral districts, many canvassers I know shifted to talking about the Democratic candidate for labor commissioner, Braxton Winston, whose proposals represented the biggest shakeup to the economic status quo, or Mo Green, a candidate for state education superintendent promising to expand free lunches. (Unlike Winston, who lost, Green and the other Democrats who won statewide races were able to complement the ground game by raising millions of dollars to run TV ads tying their opponents to a hugely unpopular Republican running for governor. Imagine if there had been ads about what the labor commissioner would do to change the working conditions for construction and food service workers?) But in cities like Richmond, Va. or Richmond, Calif., the same organizers and volunteers who earlier in the summer had been coordinating mobilizations to win commitments from elected officials for more affordable housing and to lower renter energy bills could point to local races where those issues were effectively on the ballot. In previous cycles, groups like KC Tenants have completely reshaped local races with their constant issue-based campaigning. This election wasn’t a wipeout up and down the ballot for candidates endorsed by independent political organizations, but imagine what the results might have been if more local and statewide races had been defined by our issue campaigns by the time October rolled around.

Many of us will be participating in field program debriefs over the next four months, as voter file updates come through and the results of get-out-the-vote experiments are made public. There will be more pressure on us than ever to rush that process to respond as Project 2025 is implemented. I hope we take our time and consider how we might have reached some of the constituencies that defected to MAGA Republicans had we redirected some of our capacity to campaigns designed to force Democrats to advance radical solutions to urgent problems, and cut through what Sifry describes as ​“radically different information worlds”: 

Most of the time, when people are confronted with ideas or facts that contradict their prior beliefs, they instinctively try to reinforce their pre-existing identity. Changing minds is hard and slow work. And far too much of the time and money put towards that task is spent only in the final months and weeks of an election campaign, for the relatively modest outcome of gaining votes.

If we want to build trust, change minds, and show that even Democrats can be radical proponents of upending the status quo, we can’t wait for a Bernie-like figure to emerge from a nomination process — we’ve got to lay the groundwork for change at the top by changing how we use our capacity on the ground.