How to Get a List of Registered Voters: Sources & Costs

A first-time campaign manager calls a friend on Saturday morning. The candidate announced two weeks ago. The kickoff fundraiser is Thursday. Volunteers are on the list. The friend asks the question that wakes the campaign manager up: where are you getting the voter file?

There is a long silence. The campaign manager has not thought about the voter file. The campaign manager has been thinking about the website, the kickoff, the press release. The friend says: "You can't door-knock without a list. You can't phone-bank without a list. You can't tell the volunteers who to talk to. The voter file is the campaign."

The Saturday morning that follows is the one this article exists to make less stressful. There are real options for getting a list of registered voters in 2026. They have real costs and real tradeoffs. The choice between them is not obvious, and it is not the same choice every campaign should make.

What you actually need (and what you don't)

A "voter file" is two different things wearing one name. The narrow version is the public voter registration roll — names, addresses, party (in states that record it), and registration status. The richer version is the enhanced file — the registration roll plus modeled fields layered on top: turnout history scored over the last several elections, demographic estimates, partisan support scores, issue-stance models, and contact data (phone, email) where it can be appended.

The narrow version is mostly free or cheap. The rich version is what wins close races and what costs real money. Most first-time campaigns pay for more than they need; most experienced campaigns pay for the rich version because the narrow version doesn't tell them who to spend the last weekend on.

Decide what you need before you call vendors. A school board candidate working a precinct of two thousand eight hundred voters needs different data than a state senate candidate working forty-eight thousand. The wrong answer to "which file?" usually comes from not having a precise answer to "what are we doing with it?"

Where voter files actually come from

There are five real sources in the United States, plus a sixth that is sometimes a real source and sometimes a polite trap. The five are state SOS offices, party-aligned vendors, the parties themselves, integrated platforms like NGP VAN, and consumer-data brokers offering voter overlays. The sixth is your friend's spreadsheet from the last cycle.

1. The state Secretary of State (or equivalent)

Every state maintains the registration roll, and most states sell it to qualified buyers — typically registered candidates, parties, certain committees, and (in some states) journalists or academics. Pricing ranges from free in a handful of states to several thousand dollars in others; some states limit who can buy it; most prohibit using it for commercial purposes.

The state file is the cleanest possible source: it is the actual roll, updated by the official record. It is also the thinnest. No turnout history modeling beyond what's in the public record, no scores, no phone, no email — just registration. For a small race in a state with a generous SOS, this is all you need. For anything competitive, it is a starting point, not a finish line.

2. Catalist (Democratic-aligned), L2 (nonpartisan), PDI (Republican-aligned)

These are the data shops that build enhanced files on top of the SOS roll. Catalist serves the Democratic and progressive ecosystem. L2 sells to anyone — Republicans, Democrats, nonpartisans, journalists, marketing firms. PDI is the Republican-aligned counterpart that publishes its own enhanced file. Each one layers turnout history, modeled support, demographics, and contact data — but the methodology, the consumer-data sources, and the licensing terms differ enough that the same precinct can look meaningfully different across three vendors.

Pricing is per-record and per-cycle, with steep discounts at scale. Expect a small race to pay several hundred dollars for a usable enhanced file; a state legislative race in a competitive district to pay low thousands; a congressional race to pay mid-thousands or more depending on append-data depth.

3. The party itself, or your state party

Both major parties run their own voter file infrastructure for candidates running on their line. Democratic candidates typically access the file through state parties or the Democratic National Committee's relationship with NGP VAN. Republican candidates typically access through state parties or the RNC's data trust relationship. In both cases the file is conditional on candidate registration with the party and adherence to data-handling rules.

The party file is often the cheapest path for a partisan candidate, sometimes free at the state-legislative level. It is also the most politically constrained — the data is licensed for the campaign, comes with usage restrictions, and gets clawed back when the cycle ends. Plan for the data to leave with the cycle.

4. NGP VAN, EveryAction, NationBuilder, and integrated platforms

These platforms bundle voter data with the canvassing tool. The voter file lives inside the platform; the canvassing app reads from it directly. The integration is the value proposition — no exports, no reformatting, dispositions write straight back to the file. The cost is platform lock-in: the data and the workflow live together, and switching tools at the end of the cycle is harder than switching when each layer is separate.

For a Democratic campaign with VAN access through the state party, this is often the path of least resistance. For everyone else, it is one option among several, and the integrated convenience needs to be weighed against the political and economic cost of single-vendor dependence.

5. Consumer-data brokers with voter overlays

Marketing-data firms like Experian, Acxiom, and others offer voter-registration overlays as part of broader consumer-data packages. They are not voter-file vendors first; they are consumer-data vendors who happen to license the voter roll. The data is often less current than what Catalist or L2 sells, the modeling is less politically sophisticated, and the cost can be surprisingly high because you are paying for the consumer-data platform around the voter information.

Useful when the campaign already has a marketing-data relationship and needs voter overlays added. Almost never the right starting point for a campaign whose primary need is voter contact.

How to actually choose

Walk down a short list of questions. The answers compose themselves into the right vendor.

What office? Federal congressional + statewide candidates almost always end up at Catalist or L2 (Democrats / nonpartisan) or PDI (Republicans), through a state party or directly. State legislative candidates often start with a party file and add an enhanced overlay. School board, water district, and small-municipal candidates usually buy directly from the state SOS and skip the enhancement entirely.

How competitive? Safe seats need less data; the registration roll is enough to identify likely supporters and door-knock them. Marginal seats need the modeled turnout score, the support score, and the contact data — anything less leaves persuasion targets unserved on the last weekend.

What's the in-house data capacity? A campaign with a data person who knows what to do with appended modeling will get more out of an enhanced file than one without. Buying a richer file than the campaign can analyze is one of the most consistently expensive mistakes in field operations.

What's the timeline? Enhanced files take days to license, configure, and load. SOS files take longer — some states process orders in two-to-six-week windows. If the campaign is starting today and needs doors knocked in three weeks, the timeline forces an integrated platform or a state-party file, because there's no time to negotiate a Catalist contract from scratch.

The buying mistakes that cost campaigns the cycle

Five named failures show up every cycle. Each one is the difference between a list that wins races and a list that costs money.

Buying without a target universe.

"We need the voter file" is not a buying decision. "We need a turnout-scored, partisan-affiliation-tagged file of likely 2026 primary voters in district 47, with phone append for the top 30% of support scores" is. The vendor cannot help a campaign that hasn't done the targeting math first.

Treating the enhanced file as the strategy.

Modeled scores tell the campaign who is likely to be persuadable. They do not tell the campaign what to say. Campaigns that buy the file and outsource their thinking to it are buying the wrong product. The file is an input. The strategy is still the campaign manager's job.

Forgetting the licensing terms.

Most vendor files are licensed per cycle, per campaign, and prohibit redistribution. The volunteer who downloads a copy to their personal Dropbox is in violation. The consultant who carries data across two clients is in violation. The campaign that gets reported is the one that loses access in the next cycle, which is the cycle that matters.

Buying late.

A file licensed two weeks before primary day is a file that has not been validated, deduped, or run through a turnout model the campaign trusts. Buy early. The modeling is more valuable when the campaign has time to learn what it's saying.

Skipping the comparison.

Three vendors will give a campaign three different views of the same precinct. The campaign that buys without comparing is buying without knowing what they're missing. Catalist's modeled turnout is not L2's modeled turnout, and neither is PDI's. The right answer for the race depends on which model is actually predictive in that geography for that office, which is knowable only by looking at more than one.

What to do once the file lands

Open the CSV. Look at the first hundred rows. Verify the addresses are formatted the way the canvassing tool expects (street number and street name in separate or combined columns, however the tool ingests them). Run a duplicate check on full-address strings. Run a missing-data check on the columns you actually need.

Then upload it into the field-operations layer. How to prepare the campaign once you have the list is a separate operational guide; it picks up where this article ends. The seam between buying the file and turning it into knockable routes is where most first-time campaigns lose their week.

If your race is a precinct delegate operation specifically, the six-step delegate setup covers the narrower workflow. The voter file you buy and the delegate-eligibility logic you apply to it are the two halves of that operation.

What this looks like in other verticals

The voter-file question has equivalents in every door-to-door vertical. Outside sales teams buy prospect lists from ZoomInfo, Apollo, Dun & Bradstreet, or build them in-house from public business filings. Roofing contractors chasing storm damage license property-record data from CoreLogic, Hailtrace, or county appraisal districts, with hail-impact overlays bought separately. Insurance agents marketing Medicare Advantage work from CMS-permitted lead sources only — and the compliance regime around what a Medicare prospect list can contain is far stricter than anything political.

The structural pattern is the same across verticals: there is a public underlying data source, there are commercial enhancers who layer modeling on top, and there are integrated platforms that bundle data with workflow. The choice between them comes down to the same questions every time — what are you doing with it, how competitive is the market, what's your data capacity, and what's your timeline.

Frequently Asked Questions

Can I get a voter list for free?

In a few states, yes — the SOS publishes the registration roll publicly. In most states, the file has a fee. Even in the states where it's free, you'll likely need a partisan or commercial enhancement on top to get the modeled turnout and contact data that makes the list useful.

How long does it take to get a voter file from a vendor?

State SOS files: two-to-six weeks in most states. Vendor enhanced files: typically two-to-five business days from contract signing to data delivery, faster for repeat customers. Party-state-committee files: one-to-three days for partisan candidates with credentials in order. Plan accordingly; a campaign that needs doors knocked next weekend should not be calling Catalist for the first time today.

Is L2 better than Catalist or PDI?

L2 is nonpartisan and sells to everyone, which makes it the right answer when partisan affiliation isn't the determining factor or when the campaign needs cross-spectrum data. Catalist is generally the better answer for Democratic and progressive campaigns running serious turnout operations, with deep modeling tuned to that ecosystem. PDI is the analogous answer for Republican campaigns. Better here is contextual; the right model depends on the race.

Can I share a voter file with my consultant or other campaigns?

Almost never under standard licensing. Voter files are typically licensed per-campaign, per-cycle, and prohibit redistribution. A consultant working multiple campaigns either licenses separately for each or arranges a multi-campaign license up front. Volunteers downloading the file to personal devices is a license violation in nearly every vendor's terms.

How is this different for non-political door-to-door work?

Different sources, same structure. Sales prospect data, property records, Medicare prospect lists each have their own vendor ecosystem and compliance regime. The methodology of choosing — define the universe, evaluate the modeling, plan for licensing — is identical across verticals.

Do I need a voter file if I'm running a small local race?

Almost always yes. Even a school board race in a precinct of two thousand voters benefits from knowing who has voted in the last two off-cycle elections. The narrow SOS file is usually enough; the cost is small; the alternative — knocking the universe blind — wastes weekend hours that the campaign cannot replace.

Buy the file. Then turn it into routes.

Getting the list is the first half of the operation. Turning it into a campaign that actually contacts people is the second half. Most first-time campaigns spend more energy on the first half and not enough on the second; the doors that get knocked because the routing was right are the doors that move the polling needle.

WalkLists is the field-operations layer that picks up after the voter file lands. Upload the list, draw the turf, generate the routes, hand it to your team. Start a free account — first hundred contacts free, no credit card. Or, if you're sizing for a larger team or a longer cycle, pricing is here.

For the operational setup once the file is in hand, see how to prepare a canvassing campaign. For the field-operations data discipline that determines whether the file actually delivers, see the ten field-ops data mistakes. Both are written for the campaign that just bought the file and is staring at a CSV wondering what to do next.