How to Prepare a Canvassing Campaign with Walk Lists

It's 6:47 pm on a Tuesday in late October. Maria has knocked forty-one doors. Three to go.

The route was supposed to take four hours. It took six and a half. Twelve of the houses had the wrong number — close, but the canvasser before her knocked there last week and her supervisor never updated the list. Two of the houses don't exist. One belonged to a woman who moved in 2019. At door fourteen, a man with a gun answered. He was friendly. Maria was not.

Tomorrow she'll do it again. So will twelve thousand other canvassers — and roofing crews after a hailstorm, and Medicare agents working scheduled appointments, and B2B sales reps walking commercial parks. The route is the bug. Not the people.

The waste nobody puts on the dashboard

A clean route runs at twenty-five doors an hour. A bad route runs at fifteen. Sounds small. A 100-door route at twenty-five takes four hours; at fifteen it takes six and a half, and the canvasser bails on the last twenty doors. Two hundred canvassers, six-week sprint — eighty thousand missed conversations.

For a roofing crew chasing storm damage, the same arithmetic translates to about three hundred thousand dollars of work that goes to a competitor who routed their crew better. The waste is invisible because the waste is human. A canvasser walking the wrong street doesn't show up on any chart. By the time the turnout numbers come in or the storm season closes, the money is already spent.

What's actually broken

Six things break every door-to-door operation. They're predictable enough that you can grade a campaign's setup before the first knock and forecast the result inside fifteen percent.

The list isn't geocoded — half the routes send the rep to addresses that don't resolve, and the rep wastes a minute per door figuring out where they are. The routes are too big — two-hundred-door routes that no human finishes, leaving thirty percent of the turf untouched while the manager pretends it's covered. Two reps work the same block — no central turf assignment, so blocks get double-knocked while three streets over get nothing.

The data dies in transcription — every canvasser writes notes their own way ("not home," "NH," "didn't answer," a smiley face), and what hits the spreadsheet on Monday is unusable. The re-knock list never gets out — "not home" responses sit in a notebook, then a stack on a desk, then nowhere, and warm leads cool off in three days.

Every one is a setup failure. Not a canvasser failure. Not a list failure. A thirty-minute decision somebody didn't make six weeks ago.

Same arithmetic, four ways

The numbers don't care which vertical you're in. The waste compounds the same way whether you're knocking for votes, contracts, claims, or commercial accounts. Four scenes, drawn from the four corners of door-to-door work.

The roofing crew the day after a hailstorm

Six pm Friday. A storm clipped the western edge of a metro at three. By six, fourteen contractors are already on the ground — some with storm-route software, some with a printed map and a Sharpie. The contractor who started canvassing within the first ninety minutes will sign three jobs by Sunday. The one who shows up Monday will knock the same blocks and find every door already taken.

The crew that wins the weekend isn't the crew with the best pitch. It's the crew that mapped the hail swath, pulled property records inside it, and dispatched optimized routes before the competition finished arguing about which neighborhood to start with. Speed-to-lead in roofing is a 7× win-rate multiplier. The lead is gone in twenty-four hours.

The outside sales rep in a commercial park

Nine am Tuesday. A B2B field rep has a list of three hundred small businesses across a fourteen-square-mile commercial park. With outside sales territory software and a clean route, she'll hit thirty businesses today and have meaningful conversations at twelve. Without it, she'll drive back-and-forth between clusters, hit eighteen, and have meaningful conversations at four.

The difference isn't her — she's the same rep. The difference is whether somebody routed her day. The bad route doesn't just cost the difference between thirty and eighteen. It costs the next day, too, because she goes home tired from driving instead of energized from selling.

The Medicare agent on scheduled appointments

Wednesday afternoon during AEP. A senior insurance agent has six pre-scheduled in-home appointments across the county. Medicare Advantage marketing rules mean he can only legally door-knock at the addresses on his calendar — unsolicited contact has been prohibited under 42 CFR 422.2264 since long before the 2026 rule, and CMS just expanded the plan-side accountability that surrounds it.

Routes between the six appointments determine whether his afternoon ends at six or eight pm. The audit trail of where he went and when matters more than the route, because compliance documentation is what survives the next CMS inquiry. Either you can produce the per-rep activity log, or you cannot. There is no middle ground.

The precinct canvasser for a state legislative race

Saturday morning, six weeks before a primary. A volunteer is knocking a precinct of three hundred and ninety likely supporters for a state house candidate. The voter file is current. The route is tight. She knocks ninety doors in three hours and contacts forty. The campaign has eleven volunteers like her and a six-Saturday schedule.

If the routing is right, this campaign contacts every persuadable in the district twice before primary day. If the routing is wrong, half the district never hears the candidate's name from a human. State legislative races turn on margins of forty votes; this is exactly where the routing pays for itself.

The thirty-minute setup

The first time, it takes thirty minutes. The fifth time, five.

Step 1: Get the list out of wherever it lives

Export to CSV. From your voter file vendor, your CRM, your spreadsheet, the consultant who emailed you a file two months ago — get the list into a CSV with first name, last name, street address, city, state, ZIP. Optional but useful: party, support score, phone, email, custom tags.

Save it as comma-delimited UTF-8. If the file has two hundred columns, prune to ninety. Extra columns don't help canvassers, and they slow the map.

Step 2: Upload and tell the system which column is which

Drag the file onto the upload area. WalkLists shows the first few rows in a preview grid and labels each column with what it thinks it sees — first name, ZIP, street number. The auto-detection is right about eighty-five percent of the time.

Two misses to spot before you click Continue: the system flags column A as first name when the actual first name is column C, and the address columns get half-mapped — "1234 Main St" gets tagged street number when the same column also holds the street name. Both are two-click fixes on the preview. Both cost hours later if you skip them.

Spend ninety seconds on the preview. It saves a midnight when a canvasser realizes every "Hi, John" should have been "Hi, Maria."

Step 3: Geocode — or accept the cost of skipping it

If the list doesn't have latitude and longitude, WalkLists geocodes through CSV2GEO before route generation. It's a paid add-on, priced per thousand addresses, shown before you confirm.

ZIP-centroid geocoding — what some free canvassing tools quietly fall back to — puts every address on the same dot in the middle of the ZIP. Routes built on top of that are random. CSV2GEO hits ninety-seven percent street-level accuracy on US addresses, which is the difference between Maria's Tuesday and a Tuesday that finishes on time.

Step 4: Draw the turf and generate the routes

Now the geography. Three options: draw a polygon on the map for unusual neighborhoods, pick a precinct or ZIP from the dropdown for political GOTV work, or skip the boundary entirely if your list is already filtered. Anything outside the boundary stays in your account but doesn't get routed.

Then the two sliders. Stops per route — how many doors each rep handles in a shift, default eighty. Total routes — caps the number of routes generated. Drag the sliders. The map redraws live. Tight clusters of sixty doors look very different from sprawling hundred-and-fifty-door beasts. Pick the version your team can actually finish, not the one your team would finish if they were the team you wish you had.

The TSP solver underneath picks the shortest path through each route. Typically thirty to sixty percent less driving than a human would draw freehand.

Step 5: Hand off the field package

Each route gets an assigned canvasser and a delivery format.

Mobile PWA — the canvasser opens the route on their phone, taps each stop to log status, works offline. Right for most teams. Paper — print a packet with the address list, map, and disposition checkboxes. Right for crews without smartphones, or for insurance agents working scheduled appointments where a phone-out-at-the-door look isn't right. Paper plus QR — paper packets with a QR code on every page; after the shift, snap photos and the system auto-tags each sheet back to the right route. Right when you don't trust the network.

Hand off the packets. Open the live dashboard. Watch the dispositions roll in.

The mistakes that show up in every retrospective

Field directors review the same failures campaign after campaign. Naming them costs nothing; making them costs the campaign.

Letting the rep do the routing in the field.

A canvasser who's been knocking the same neighborhood for years thinks they know the optimal path. They don't. They know the path that feels productive, which is the path that ends near their car. The TSP solver doesn't care where their car is. It cares where the doors are.

Treating turnout / closing rate as the canvasser's metric.

Turnout is a system metric. Closing rate is a system metric. Both reflect the routing, the list, the timing, the script, and the season — not the rep's effort. Reps who get reviewed on system metrics burn out. Reps who get reviewed on contact rate, dispositions logged, and route completion stay.

Underestimating the audit trail.

The agency that wins the next contract is the one that can produce the per-rep activity log in twenty-four hours. The campaign that survives the post-cycle FEC inquiry is the one whose disposition records reconcile with their volunteer hours. The crew that wins the next storm-chase deal is the one whose claim documentation backs up every door visit. The audit trail isn't paranoia — it's the next sale.

Six habits that separate field directors who run canvasses well from the ones who run them once

Knock 9–11 am and 5–7 pm. People are home. Dogs are calmer. Lunch and dinner are dead time — give the canvassers a real break instead of pretending they're knocking.

Cap routes to your slowest canvasser, not your best. A 100-door route at twenty-five an hour assumes a confident veteran. Plan for the canvasser you actually have.

Pre-print QR-tagged paper backups even when you're going digital. Phones die. Cell signal drops in storm zones. A paper sheet with a QR code never crashes.

Brief the team on the survey questions before the shift. Five canvassers asking the same question five different ways equals unusable data. A two-minute briefing fixes it for the entire campaign.

Send the re-knock list out tomorrow, not next week. Two-day-old "not home" leads convert about three times higher than week-old ones. Warm contact cools fast.

Walk the first route yourself. You'll catch every problem your team will hit, before they hit it. Worst hour you'll spend that week. Best hour for the campaign.

Frequently Asked Questions

How many contacts can a single walk list hold?

There's no hard cap, but routes over about a hundred and fifty doors get unwieldy and rarely finish. Most teams settle on eighty to a hundred and twenty doors per four-hour shift.

Can I run a canvass without geocoding?

Only if the CSV already has latitude and longitude. Routes need coordinates to exist. Files without them get geocoded through CSV2GEO before route generation — the difference between routes that work and routes that send people to the middle of a ZIP code.

What's the difference between a walk list and a campaign?

A campaign is the whole effort: list, turf, team, time window, dispositions. A walk list is one route handed to one canvasser. A typical campaign has five to fifty walk lists.

Does WalkLists work for non-political teams?

Yes. The same engine runs outside sales territories, roofing crew dispatch in storm zones, Medicare Advantage appointment routes, and political precinct walking. The mechanics of door-to-door don't change much across verticals; the data and the compliance regime do.

What happens to contacts the canvasser never reaches?

They flag automatically as not-attempted in the disposition log and roll forward to the next route generation. The list doesn't lose them; the routing system rebalances. This is why the re-knock list works without being manually rebuilt — the data flows from this round into the next one.

Is paper or digital better for canvassing?

Digital wins on data quality. Paper wins on reliability in dead-cellular zones. The hybrid model — paper packets with QR codes that auto-tag back to the digital system after the shift — wins on both. Most teams running a serious operation use the hybrid.

How is this different from the old paper-map way?

Three things. Routes are TSP-optimized so the canvasser walks less. Every disposition feeds a real-time dashboard so the manager sees coverage as it happens. Re-knock lists generate themselves from the responses, so warm leads don't cool off in a notebook. Paper maps give you addresses. WalkLists is the operations layer on top.

Run the math on your next canvass

Most field directors who switch say the same thing on the call after their first campaign: I had no idea how much we were leaving on the table.

Free for the first hundred contacts, no credit card, start an account. Upload the list, draw the turf, generate the routes, hand it to your team — in less time than it takes to print the old paper packets. If you're sizing for a larger team or running a longer cycle, pricing is here.

For voter list buying specifically, there's a separate guide. For precinct delegate operations, the six-step setup is more specific. Both are worth the read before you start.