Here’s how Zohran Mamdani’s 50K-strong volunteer army pulled it off

It wasn’t just the TikTok videos.

Zohran Mamdani’s primary upset was powered by an army of tens of thousands of volunteers who knocked on doors and phonebanked for him. The campaign’s field operation was a scaled-up version of a tactic that the New York City chapter of the Democratic Socialists of America (which I’m a former member of) has been honing for the past several years.

Much of the credit for Mamdani’s unprecedented canvassing operation belongs to Tascha Van Auken, the campaign’s field director. Van Auken has years of experience with these kinds of field operations. She worked as a field lead on former President Barack Obama’s 2008 campaign before managing Julia Salazar’s successful 2018 state Senate campaign and Phara Souffrant Forrest’s successful 2020 Assembly campaign (both of which were backed by DSA) and then served as the New York Working Families Party’s deputy campaigns director during the 2022 cycle. In between all the campaigns, she worked as a casting director and artistic direction manager for the Blue Man Group.

According to Van Auken, more than 50,000 people signed up to volunteer for Mamdani’s campaign – and more than 30,000 of those volunteers worked as canvassers, knocking doors or phonebanking. (The campaign also hired 40 to 50 specialized paid canvassers, largely to reach voters in less accessible areas of the city and to reach voters who spoke languages that few volunteers were familiar with.) The Mamdani campaign’s volunteers knocked on doors 1.6 million times, which led to 247,000 conversations with voters at their doors. To put that in perspective, Mamdani’s field operation ended up speaking to about a quarter of the total number of people who voted in the mayoral primary. 

City & State spoke to Van Auken about how a DSA-style volunteer field operation actually works and what it takes to manage tens of thousands of volunteers. This interview has been edited for length and clarity.

For those unfamiliar with DSA’s model of canvassing, can you explain how the campaign’s large-scale field operation actually worked?

We started very early in December, and the reason that I think it's important to start early – and I think it was definitely very important for us – was we were building the model for what we were doing, and we were developing leadership. You can't just scale up overnight. People need to be trained. People need to see what it is we're doing, learn how to do it, step into those roles. We need to troubleshoot. We start with a script. We tweak the script based on feedback that we get, that kind of thing. So I would say that the beginning of the campaign for us is building up the leadership structure, building up the infrastructure, really honing in on what we're talking to voters about – which on this campaign was pretty easy and straightforward, because the message is one that just resonates with so many people. 

Then that leads us into petitioning, which lasts a little over a month, which is where we sort of shift away from doors to – we do some doors, we do a lot of street petitioning. Then once petitioning is over, we go full force into persuasion, which is basically just lots and lots of talking to voters that we haven't talked to before, persuading them to vote for Zohran or just identifying that they are Zohran supporters or not. Then the last few weeks, we add, I say add, because early voting means that we're doing persuasion and get out the vote at the same time. But the last few weeks we add get out the vote, which is just, we're recontacting people who are supporters, making sure they vote. 

When did you stop doing persuasion?

Never. People really start to tune in, a lot of voters only start to tune in in the last couple of weeks. So it doesn't really make sense to completely stop doing persuasion in the last couple of weeks when we’re doing (get out the vote), so we do both at the same time.

Canvassers will often speak about “ones and twos,” which is a reference to the scale that is used to rank voters’ support for a specific candidate. Can you explain how the scale works?

We tweaked (the scale) a little bit for rank choice to identify folks who would vote for Zohran, but not rank him No. 1, because those were important voters, and we wanted them to vote as well.

Typically when we do a 1 through 5 scale, when it's just straightforward, 1 is like, “I've heard of this person already. I support them” and 2 is like, “Never heard of them, but support them.” So we did the same thing. 1: “Strong support. Heard of them.” 2A: “Haven't heard of them. Like them, will rank them first” 2B: “Like Zohran, going to rank Zohran, but not going to rank him first.” And then 3, 4 and 5 are all the same (as the normal scale). So 3 is “undecided,” 4 is “leaning against” and 5 is “strong oppose.”

Then when you do get out the vote, the idea is to only go and recontact the ones and twos to remind them to vote, right? You’re not recontacting the fours and fives.

No, no, no, no, of course not!

Was field involved in fundraising for Mamdani at all? Before he reached the spending cap, were canvassers knocking on doors and asking people to donate?

Early on, we did that a bit, and it was surprisingly successful. I've tried to do fundraising on past campaigns. It's a very hard thing. Anybody who's done fundraising canvassing at doors will tell you that's like the hardest canvassing you could do. But we added it early on, and we actually would get quite a few people who would donate on the spot. But we stopped once we hit the goal.

You mentioned that you can’t scale up a field operation of this size overnight. What did the structure of the field operation look like?

We had, ultimately, over 400 field leads, but that's over the course of the entire campaign. We would have trainings for our field leads, where we would train them on how and what you're doing. 

I think the thing that is very, very important for something of this size – a field operation of this size and this many volunteers doing this much work – is that the candidate and the message and what we're fighting for has to be something that people believe in and really care about. So we talk about that, and we talk a lot about why we're there. People were coming into this campaign for very, very personal reasons. Part of the training as well is just really understanding why this election is so important, why this campaign is so important, why what we're fighting for is so important.

I think ultimately, going back to December, having that model of what it is we're doing and that first batch of people who are able to lead canvasses. Many of those folks in December, it was a “DSA for Zohran” launch, came from DSA and had led canvasses before. Ultimately, I believe, people learn the best by doing a thing or watching someone do it. I think everything is about replication and really learning from one another, and if you're a leader of some kind, looking for other folks who can also do the same thing you're doing and teach them how to do it.

One difference between a volunteer field operation like this and a professional or paid canvassing operation is that you’re not just asking people to vote for Mamdani but to volunteer for him. How does that volunteer pipeline work?

Aside from the fundraising ask, I would say the second-hardest ask at a door in past campaigns I’ve done is: “Will you volunteer for the campaign?” Generally, you include that after you find out somebody is a supporter. You talk about what it is you're doing and ask them if they would like to participate and join the campaign. Typically, it's a hard ask. I would say that was another one on this campaign where we had lots and lots of people from doors wanting to volunteer. Again, I think that really speaks to the urgency of the message and the moment. So once somebody says they want to volunteer, we follow up with them really, really quickly and ask them to canvass or ask them to phonebank. We train all of our field leads to basically identify anyone who's really enthusiastic and wants to do more and then bring them in as a field lead. We just have a leadership model where we're trying to replicate leadership on the campaign.

Would you say you’re democratizing the field operation?

Exactly. I think we also learn a lot (from) the field work. Because it's at this scale and because we have people who really care, we learn a lot from the conversations at doors. We learn really fast what is resonating with voters and what isn't. There is this sort of feedback from voters directly that affects the campaign and what we're saying.

What do you have planned for the general election?

I don't have much information to give you at this moment, but we will be running a serious general election, and activities will start up soon.