Roofing Canvassing Software That Targets the Right Houses, Not More Doors

A canvasser knocks roughly 80 to 120 doors on a good day. Maybe 25 people answer. Maybe 6 of those will let you on the roof. Maybe 2 of those roofs are actually worth replacing right now. Everything else is windshield time, polite no-thank-yous, and a crew that goes home tired with nothing to show the production board.

That ratio is the whole game. And here is the uncomfortable part: most roofing canvassing software does almost nothing to improve it. The typical tool draws a polygon on a map, lets you drop pins, and tracks who knocked where. Useful for accountability. Useless for the only question that determines whether your canvass pays for itself: which of these houses has a roof that's actually due?

Knocking more doors is not a strategy. It's a tax you pay because you don't know which doors matter. The contractors who win in a crowded storm market aren't the ones with the most reps on the street. They're the ones whose reps spend their hours in front of the right houses. This is a long look at how to choose canvassing software that targets the right houses, how to build a field workflow around it, and the operational details that separate a canvass that prints money from one that just burns gas.

We'll cover what canvassing software actually does (and where the category quietly oversells), how roof-age and storm-physics targeting changes your knock list, how to plan territories and routes that don't waste a crew's legs, the metrics that tell you whether any of it is working, the compliance landmines (DNC, permits, hail-chasing reputation), and the edge cases nobody warns you about. There's a worked example with real numbers, a few checklists you can hand a crew lead tomorrow, and an honest section on where the data falls short.

What Roofing Canvassing Software Actually Does

Strip away the marketing and most canvassing platforms in roofing do four things:

1. Map and territory drawing. You outline neighborhoods, assign them to reps, and color-code who owns what. This prevents two of your guys from knocking the same street and getting into a turf fight on a homeowner's porch.
2. Pin dropping and disposition tracking. Every door gets a status: not home, not interested, inspection set, signed, callback. The rep taps a button, the office sees it live.
3. Lead capture and CRM handoff. When someone says yes, the rep grabs contact info, maybe a photo, and pushes it into your CRM or production pipeline.
4. Activity reporting. Doors knocked per rep, contact rate, sets per hundred knocks, conversion. Manager dashboards, leaderboards, the works.

All four are genuinely useful. A crew running paper and memory loses leads, double-knocks, and can't be coached because there's no data. So the baseline value of any canvassing app over a clipboard is real.

But notice what's missing from that list: none of it tells you which houses to knock in the first place. The software assumes you already know your territory is worth canvassing. It optimizes execution. It does nothing for targeting. You draw the polygon; the tool counts the doors inside it. Whether those doors hide 2-year-old roofs or 28-year-old roofs, whether the last storm clipped that subdivision or skipped it entirely, the software is blind to all of it.

That blindness is the single biggest reason canvassing budgets get torched. You can have flawless route discipline and a beautiful disposition funnel and still lose money, because you spent the whole week knocking a neighborhood full of roofs that won't need replacing for a decade.

The two questions targeting has to answer

Good targeting collapses to two questions, asked house by house:

• Is this roof aging out? Asphalt shingle roofs in most of the country have a practical service life in the 15-to-25-year range depending on product, ventilation, slope, and climate. A roof in that window is a candidate on age alone, storm or no storm.
• Did weather actually wear this roof out faster? Hail, straight-line wind, and repeated thermal cycling don't hit a neighborhood evenly. The same storm can shred one street and leave the next block untouched, because hail swaths are narrow and wind funnels around terrain and structures.

A house that scores high on both, an older roof that also sat under a verified hail core, is the door you want your best closer knocking first. A house that scores low on both is windshield time. Most software can't tell these two houses apart. The good news is that the data to tell them apart now exists, and that's where the category is moving.

Why "More Doors" Is the Wrong North Star

Let's do the math that most sales managers avoid because it's depressing.

Say a rep knocks 100 doors. Industry-typical contact rates (someone actually answers) run 25 to 35 percent in suburban single-family neighborhoods, lower in areas with a lot of renters or daytime-empty homes. Call it 30 contacts. Of those, a strong rep might set inspections on 15 to 20 percent, so 4 to 6 inspections. Of inspected roofs, the share that genuinely warrant replacement depends entirely on whether you targeted well. In a random neighborhood with no storm and no age screen, that might be 1 in 4 inspections. In a well-targeted older neighborhood that took a verified hail hit, it can be more than half.

Here's the same 100 doors two ways:

Same labor. Roughly 2.5x to 3x the signed jobs, because the houses behind the doors were different. The rep didn't get better. The list got better. That's the entire argument for targeting software, and it's why "doors per day" as a north-star metric quietly rewards the wrong behavior. A rep optimizing for door count will speed-knock a cheap, dense, low-value neighborhood and post huge activity numbers with nothing to show downstream.

What "more doors" actually costs

Every door has a cost even when nobody answers: a few minutes of a paid rep's time, fuel, vehicle wear, and the opportunity cost of not being at a better door. If a canvasser is fully loaded at, say, mid-teens dollars per hour plus vehicle and overhead, and knocks 100 doors in a shift, you're spending real money per door regardless of outcome. Cut the dead doors out of the route and you've effectively given every rep a raise in productivity without touching their pay. That's the prize.

How Roof-Age and Storm-Physics Targeting Changes the Knock List

This is the part of the category that's actually new, and where a tool can change your economics instead of just organizing your activity.

Two data layers matter, and they're different:

Roof age, expressed as a range. You cannot get a precise install date for a stranger's roof from the curb or from the sky. Anyone who promises an exact year is selling you certainty that doesn't exist. What modern aerial-imagery analysis can do is estimate a roof's age as a range, by comparing historical imagery (when did the roof last visibly change?), reading surface condition cues, and factoring material type and regional norms. A range like "roughly 14 to 19 years" is honest and, crucially, actionable: it's enough to sort a street into "likely aging out" versus "probably has years left." Treat any age figure as a range, never a date.

Storm exposure, modeled per roof. Hail and wind data at the ZIP-code or county level is nearly worthless for canvassing, because the damage that matters happens at the scale of a single street or a single roof. What helps is storm modeling resolved down to the individual structure: given the hail size, wind speed, direction, and the roof's slope, orientation, age, and material, how likely is this specific roof to have suffered functional damage? That's a probability, an estimate of odds, not proof of damage. Proof still comes from a physical inspection. But odds are exactly what you need to decide who to knock first.

This is the slot where a tool like RoofPredict fits. Rather than handing a crew a polygon and a door count, it ranks the houses inside a territory by which roofs are most likely due, combining a per-address roof-age range from aerial imagery with storm exposure modeled per individual roof. The output isn't "here's a neighborhood," it's "here's the order to knock this street, oldest-and-most-storm-worn first." A crew leader can hand the highest-probability doors to the strongest closer and route the rest efficiently instead of knocking north-to-south by accident of geography.

What it honestly does not do is tell you a roof is definitely damaged, give you an exact install date, or replace the inspection. It narrows where to look and what order to look in. The roofer still documents actual conditions on the roof, writes the estimate, and lets the homeowner and their insurer sort out coverage. The software's job ends at getting your best person to the right ladder. That boundary matters both legally and for trust: you're prioritizing attention, not pre-judging claims.

Building the targeted knock list, step by step

Here's how the targeting layer turns into a route a rep can actually run:

1. Define the candidate area. Start with where a storm hit, or where your install-history and aging data say roofs are coming due. Don't start with "a neighborhood that looks nice."
2. Pull the per-roof scores. For every address in the area, get the roof-age range and the per-roof storm-exposure odds.
3. Sort into tiers. Tier A: older roof range and high storm odds. Tier B: one or the other. Tier C: young roof, no storm exposure, skip for now.
4. Sequence the route by tier, then by geography. You want to hit Tier A doors first, but not by zig-zagging across the map. Cluster Tier A houses geographically so the route is both high-value and efficient.
5. Assign by skill. Tier A doors to your closers. Tier B doors to developing reps for at-bats. Tier C left alone or saved for slow days.
6. Push to the field app so the rep sees the order, the why, and can disposition each door as they go.

That sequence is the difference between "a list of houses" and "a plan for the day."

Territory Management: Carving Maps That Don't Waste Legs

Targeting tells you which houses. Territory management decides who knocks them and how the work is divided so nobody steps on anybody and no street gets knocked twice. This is where canvassing software earns its keep on the operations side.

Right-sizing a territory

A territory that's too big means reps cherry-pick the easy streets and leave the rest cold. Too small and they run out of doors and start idling. A practical unit is what one rep can meaningfully cover in a focused block of canvassing time, accounting for walk time between doors and the contact-rate reality that most doors don't answer. Density matters enormously: a tight subdivision with houses 40 feet apart supports far more doors per hour than a rural road with quarter-mile driveways.

Rules of thumb that hold up in the field:

• Size territories by knockable doors, not by acreage or street miles. A square mile of half-acre lots is a totally different job than a square mile of townhomes.
• Keep territories contiguous and walkable. Splitting a rep across a highway or a river guarantees wasted drive time.
• Re-cut after a storm. Storm response should redraw territories around the actual damage swath, not around your old comfortable boundaries. The good targeting tools will show you the swath; cut to it.

Avoiding the overlap and the cold zones

The classic canvassing failure modes are overlap (two reps work the same block, annoy the same homeowners, and argue over a set) and cold zones (a chunk of a hot neighborhood never gets touched because it fell between two territories). Software solves both by making ownership explicit and coverage visible. A good territory view shows you, in color, which doors have been knocked, which are pending, and which are unassigned. A manager glancing at that map should be able to spot a cold zone in seconds.

The accountability layer

Territory software also creates a record. When a rep claims they "knocked the whole street," the pin data either backs that up or it doesn't. This isn't about distrust; it's about coaching. A rep with high doors and low sets has an approach problem. A rep with low doors and high sets is either lucky or working a great list, and you want to know which. You can't have either conversation without the data, and you can't trust the data if reps can fake-pin from the truck. The better apps timestamp and geotag dispositions so the activity is real.

Route Planning and Sequencing

A rep's day is mostly between doors, not at them. Shaving walk and drive time is pure margin. Route planning inside canvassing software ranges from "here are your pins, good luck" to genuinely optimized sequencing. The features worth paying for:

• Tier-aware sequencing. As covered above, you want high-value doors first, but ordered to minimize backtracking. A route that sends a rep to the single best house across town first and then crawls back is worse than one that works the best cluster in order.
• Time-of-day logic. Contact rates swing hard by hour. Late afternoon into early evening (until a reasonable, neighborhood-appropriate cutoff) is prime for catching people home. A smart route front-loads lower-contact-likelihood streets earlier and saves the best streets for when people are home.
• Re-knock scheduling. "Not home" is not "no." A door that didn't answer at 2pm on Tuesday should resurface at 6pm on Thursday. Software that automatically queues re-knocks at different times of day will dig more sets out of the same territory than any amount of new-door pounding.
• Live re-routing. When a rep sets an inspection or runs long on a porch conversation, the rest of the route should adjust rather than leaving them staring at a stale list.

A simple re-knock cadence

Here's a re-knock rule a crew can follow without thinking:

Three well-timed attempts catch most catchable households. Past that you're spending good time on a door that's telling you something.

The Field Workflow That Turns Targeted Doors Into Jobs

Great targeting and a clean route still fall apart at the porch if the rep's workflow is sloppy. Here's the end-to-end loop the best crews run, with the software doing the heavy lifting where it should.

1. Pre-shift huddle. Crew lead reviews the day's tiers and territory. Everyone knows their streets, their door order, and the one-line why behind the top houses ("these roofs are in the aging-out range and sat under last month's hail core"). The why matters because a rep who knows the house is a real candidate knocks with more conviction.

2. The knock and the opener. The opener is not a pitch; it's a reason for the conversation. Specific beats generic. "We've been working roofs in the neighborhood that took the hail last month, and yours is one we wanted to take a look at" lands far better than "We're offering free inspections." Targeting makes the specific opener true, which is why it converts.

3. Disposition every door. Win, lose, or no-answer, the rep taps the status before walking to the next house. Skipping this is how data rots. The discipline is non-negotiable.

4. Set the inspection, capture the lead. On a yes, the rep grabs contact info, the best inspection window, and any notes (gate code, dog, homeowner's stated concern). The software pushes this straight into the pipeline so the office isn't re-keying.

5. The inspection. This is where odds become evidence. The roofer gets on the roof (safely, more on that below), documents actual conditions, photographs everything, and writes an honest assessment. If there's no real damage and the roof has life left, say so. Burning trust on a marginal roof costs you the three referrals that house would have sent you.

6. Documentation and handoff. Clean photos, measurements, and notes go into the file. The homeowner gets a clear picture of their roof's condition and a straight estimate. From here, coverage questions are between the homeowner and their insurer; the roofer's role is to document what's there and quote the work. Keep that line bright.

7. Follow-up. Most sets don't close on the first visit. A disciplined follow-up cadence (timely, helpful, not pushy) recovers a big share of "thinking about it" homeowners. Log it all so nobody falls through.

Pre-shift checklist (hand this to a crew lead)

• Territory assigned and loaded on every rep's device
• Door order sorted by tier, clustered for walking
• Top-house "why" understood by each rep
• Re-knock queue from prior days merged into today's route
• Ladders, safety gear, and inspection tools accounted for
• Door hangers / leave-behinds stocked
• DNC and permit status confirmed for the area (see compliance below)

Metrics That Tell You If It's Working

If you only track doors knocked, you'll optimize for the wrong thing. Here's the metric stack that actually reflects whether your canvass is healthy, from top of funnel to cash.

The one most teams ignore is due rate, the share of inspected roofs that genuinely warranted replacement. It's the direct scoreboard for your targeting. If due rate is low, your list is bad, no matter how good the reps are. If due rate is high but close rate is low, your list is great and your sales process is the problem. Splitting these two apart is the single most clarifying thing you can do for a canvassing operation, and you can't do it without disposition discipline plus an honest call on each inspection.

A worked example

A contractor runs a 4-rep crew for one week, five canvassing shifts each. Numbers from the field:

• 4 reps x 5 shifts x 100 doors = 2,000 doors
• Contact rate 30% = 600 conversations
• Set rate 18% = 108 inspections
• Untargeted due rate ~26% vs targeted due rate ~55%

Untargeted: 108 x 0.26 = ~28 due roofs. Close at 40% = ~11 jobs. Targeted: 108 x 0.55 = ~59 due roofs. Close at 40% = ~24 jobs.

Same crew, same week, same effort. The targeting layer more than doubled signed jobs by changing which doors got knocked, not by working anyone harder. Even if the tool costs real money per month, the math isn't close: a single extra job typically covers the software many times over, and here we're talking about roughly a dozen extra jobs in one week. That's the case for paying for targeting instead of just paying for a door-tracker.

Choosing the Software: A Buyer's Checklist

When you evaluate canvassing platforms, separate the table-stakes execution features from the targeting features that actually move your due rate. Score each tool on both.

Execution features (table stakes):

• Territory drawing, assignment, and conflict prevention
• Geotagged, timestamped dispositions (no truck-pinning)
• Mobile-first, works offline in dead zones, syncs later
• Re-knock scheduling with time-of-day awareness
• CRM / production-pipeline integration (no double data entry)
• Manager dashboards: contact rate, set rate, and conversion, beyond raw door counts
• Door-hanger and leave-behind tracking

Targeting features (the differentiators):

• Per-address roof-age estimate, expressed as a range (be suspicious of exact-date claims)
• Storm exposure modeled per individual roof, not ZIP-level
• A ranked knock list / tiering you can route against
• Honest probability language (odds, not "this roof is damaged")
• Clear scope: prioritizes attention, doesn't pre-judge claims or coverage
• Coverage in your actual storm and service area

Questions to ask every vendor:

1. Is your roof-age output a range or a single date? (Range is the honest answer.)
2. At what resolution is storm exposure modeled, ZIP, parcel, or individual roof?
3. How do you express damage likelihood, and do you ever claim a roof is damaged without an inspection? (They shouldn't.)
4. Can I export a tiered, route-ready list to my field app?
5. How current is your aerial imagery and storm data?
6. What's your coverage in my markets?
7. How do you handle my data and my homeowners' data?

A vendor that hedges on questions 1 and 3, claiming precise dates or guaranteeing damage, is overselling. The honest tools draw bright lines around what the data can and can't prove, and those are the ones that won't get you in trouble.

Compliance and Reputation: The Stuff That Bites Later

Canvassing lives in a regulatory and reputational minefield. Software helps, but only if you've configured it around the rules. Get this wrong and you trade a short-term sales bump for fines and a community that hates you.

Do-Not-Knock, permits, and local ordinances

Many municipalities require a solicitation permit for door-to-door sales, and many neighborhoods or HOAs maintain no-soliciting registries or signage that carry legal weight. Some cities run their own do-not-knock lists. Your software should let you flag and exclude these addresses so reps never knock them. Before a campaign, check the local clerk's office for permit and registration requirements; they vary widely by city and even change after a major storm as towns crack down on out-of-area contractors.

DNC, calls, and texts

Door-knocking itself isn't governed by the federal Do-Not-Call rules, but the moment your follow-up moves to phone or text, the FTC's Telemarketing rules and the Do Not Call Registry come into play, along with consent requirements for texting. Don't let a clean canvass create a dirty call list. Capture consent properly and respect opt-outs.

The hail-chaser reputation problem

After big storms, communities get flooded with out-of-area crews, some of them sloppy or worse. State insurance and consumer-protection agencies routinely warn homeowners about storm-chasing contractors, high-pressure tactics, and "free roof" pitches. You do not want to be lumped in with that crowd. Targeting actually helps here: when you knock specific, genuinely-affected houses with a true, specific reason and document conditions honestly, you read as the local pro, not the chaser. Avoid pressure scripts, avoid promising outcomes you can't control, and never frame a storm model as proof of damage. The model says odds; the inspection says what's actually there.

Keep the claims line bright

The roofer's job is to document actual roof conditions and provide an estimate for the work. Whether damage is covered, and to what extent, is a determination for the homeowner and their insurer. Don't position yourself as adjusting, approving, or guaranteeing claims, and don't make promises about deductibles or "free" anything. Stay in your lane and you stay both compliant and credible. The targeting data supports this posture perfectly: it tells you where to look, never what a claim will do.

Roof Safety: The Part That Can End Your Business

Targeting gets your rep to the right ladder. What happens at the ladder can end careers. Falls are consistently among the deadliest hazards in construction, and roofing crews are squarely in the line of fire. Inspections are not exempt just because nobody's tearing off shingles yet.

Basic non-negotiables for anyone getting on a roof:

• Proper fall protection per the work being done; don't treat "just an inspection" as a free pass to skip it.
• Ladder set at the correct angle, secured, extending above the eave.
• No roofs in rain, ice, high wind, or when surfaces are slick.
• Steep or fragile roofs documented from the ground, drone, or imagery rather than walked when conditions are dangerous.

This is where aerial imagery earns extra credit beyond targeting: for a steep or storm-compromised roof, a lot of the documentation can happen without putting a body in a dangerous spot. The safest inspection is sometimes the one you don't have to climb for.

Edge Cases and What Pros Get Wrong

The theory is clean. The field is messy. Here are the situations that trip up even experienced canvassing operations.

Newer roofs in storm zones. A roof replaced two years ago can still take real hail damage. Don't let an age screen alone make you skip a young roof that sat under a severe core, and don't let storm odds alone make you knock a 3-year-old roof with no exposure. The two layers work together; using either one alone produces blind spots.

Re-roofs and layovers. Some roofs have a second layer of shingles laid over the first. Aerial age estimates can read the visible surface as newer than the structure really is, and a layover changes the replacement conversation entirely. Treat age as a starting hypothesis the inspection confirms.

Mixed-material and complex roofs. Tile, metal, slate, and flat sections age and fail on completely different curves than asphalt shingle. A targeting model tuned for shingle may not say much useful about a tile roof. Know what your tool is good at and where it's guessing.

The "nice neighborhood" trap. Reps love knocking affluent areas because the houses are pretty and the conversations are pleasant. But pretty doesn't mean due. A neighborhood of 6-year-old roofs is a beautiful waste of a shift. Target on roof condition and storm exposure, not on curb appeal or assumed ability to pay.

Renter-heavy areas. High renter density tanks your contact-and-set economics because the person at the door usually can't authorize a roof. Your territory tool should let you down-weight or skip rental-heavy blocks.

Over-knocking after a storm. When everyone descends on the same swath, homeowners get knock-fatigued and hostile. Being early and specific beats being loud and late. Targeting helps you get to the genuinely-affected houses before the herd, and skip the over-canvassed streets that are now a wall of no-soliciting signs.

Treating storm odds as proof. The single most common and most dangerous mistake. A high storm-exposure score is a reason to inspect, full stop. It is not evidence of damage, not a basis for a claim, and not something to wave at a homeowner as a guarantee. Every promise you make on odds is a promise the inspection might not keep. Let the roof tell the truth.

Letting data discipline slide. The whole metric stack collapses if reps don't disposition honestly. "I'll update it later" becomes never. Build the habit that the door isn't done until it's dispositioned, and audit it.

Putting It Together: A 30-Day Rollout

If you're moving from clipboard-or-basic-app to targeted canvassing, here's a sane sequence that doesn't blow up your season.

Week 1, baseline. Keep doing what you do, but start logging the real metric stack: contact rate, set rate, due rate, close rate, cost per job. You need a before-picture to know if anything's working.

Week 2, layer in targeting. Bring in roof-age and per-roof storm data for one or two pilot territories. Build tiered knock lists. Run one crew on targeted lists and one on your old approach, same effort, and compare due rate and close rate at week's end.

Week 3, tighten the workflow. Standardize the porch opener around the specific, true reason the house was targeted. Lock disposition discipline. Turn on re-knock scheduling. Fix the worst data leak you found in week 1.

Week 4, scale what worked. Move more crews onto the targeted approach, re-cut territories around real storm swaths and aging clusters, and set due rate and cost-per-job as the metrics your leaderboard actually rewards, not raw door count.

At the end of the month you'll have a defensible answer to the only question that matters: did targeting the right houses lower your cost per signed job? If your data discipline held, the answer is almost always yes, and now you can see exactly why.

Integrating Targeting Data With the Tools You Already Run

Most roofing operations already run a stack: a CRM, an estimating or measurement tool, a production scheduler, maybe a separate texting platform. Targeting data has to flow through that stack without creating a second system your office has to babysit. Here's how the pieces should connect so nobody's copying addresses by hand.

Targeting feeds the canvassing app. The ranked, tiered knock list is the input to your field tool. Reps see the order and the why; they never see a wall of undifferentiated pins. If your canvassing app can import a prioritized list (by address, with a tier and a short reason), the integration is mostly solved at the front end.

The canvassing app feeds the CRM. When a rep sets an inspection, the lead, contact info, address, tier, and notes should land in the CRM as a new opportunity automatically. Re-keying is where leads die: a rep sets six inspections, the office is slammed, two never make it into the system, and you've paid for canvassing you'll never collect on. Insist on a real integration here, not a nightly CSV someone forgets to upload.

The CRM feeds production and follow-up. Once a job is signed, it moves into scheduling and production. Once an inspection is a maybe, it moves into a follow-up cadence. The targeting data can even ride along: knowing a deferred lead's roof is deep into its aging-out range tells your follow-up which homeowners to nudge harder over the coming months.

The practical test for any vendor: ask for a live demo of an address flowing from targeting list, to field set, to CRM opportunity, to follow-up, without anyone typing the address twice. If they can't show it, you'll be the integration, and that work has a way of not getting done in the field.

When to bring data in-house versus buy it

Some larger operations ask whether they should build their own targeting from raw imagery and storm feeds. For nearly everyone, the answer is buy, not build. Turning aerial imagery into honest roof-age ranges and storm feeds into per-roof odds is a hard modeling problem that takes specialized data, ongoing imagery refreshes, and constant validation. The cost of getting it slightly wrong, sending crews to the wrong houses, isn't worth the in-house engineering. Spend your energy on the field execution that's genuinely yours to own: the opener, the inspection quality, the follow-up, the crew culture. Let the targeting be a feed you plug in.

Coaching Reps With the Data You're Now Collecting

Targeting changes what good coaching looks like, because for the first time you can separate "bad list" from "bad rep." Before targeting, a rep's whole funnel was a black box: low sets could mean weak streets or weak skills and you couldn't tell which. Now you can.

Run this read on every rep monthly:

• High set rate, high due rate, low close rate. The rep is getting in front of good roofs but losing them at the kitchen table. Coach pricing presentation, objection handling, and follow-up. The list is fine; the sale is the gap.
• Low set rate, high due rate. Great list, weak porch. The rep is walking away from good roofs because the opener or rapport falls flat. Ride along, fix the approach.
• High set rate, low due rate. The rep is charming homeowners into inspections on roofs that aren't due, or working a weak list. Check whether they're knocking off-tier; if not, the list needs another look.
• Low everything. Either a genuinely cold territory or a rep who needs fundamentals. The tier data tells you which.

The leaderboard should reward the metrics tied to cash, signed jobs and cost per job, not raw activity. The moment your board rewards doors knocked, your best gamers will speed-knock dense low-value blocks to top it, and your real producers who work fewer, better doors will look worse. Reward the outcome, and the activity sorts itself out.

A short script note

The single highest-leverage coaching change targeting enables is the opener. A generic opener ("we're doing free inspections in the area") invites the reflexive no. A targeted opener is specific and true, and specificity is what makes a stranger pause. Train reps to lead with the real reason the house was on the list, the storm the neighborhood took or the roof's visible age, then ask for a look, not for a sale. The set comes from earning two minutes of attention, and a true specific reason earns it far better than a discount pitch.

Seasonality: Targeting Outside of Storm Season

Storm chasing is feast or famine. The crews that survive the lean months are the ones that don't depend on storms for every job. This is where the roof-age layer earns its keep independent of weather. Even with zero recent storms, a market always contains roofs quietly aging out of their service life, and those homeowners need replacements whether or not hail came through.

A storm-free canvass is simply targeting on the age layer alone: find the clusters of roofs deep in the aging-out range, knock those, and lead with condition and longevity rather than storm damage. The conversation is different, it's about a roof at the end of its life and getting ahead of leaks rather than reacting to an event, but the underlying motion is identical: get your best person to the roofs most likely to be due. Operations that work the age layer in the off-season keep crews busy, keep cash flowing, and aren't desperate when the next storm hits. Desperation is what produces the high-pressure tactics that wreck reputations; a steady off-season pipeline is the cure.

Think of it as two engines running off the same targeting data:

Most contractors over-index on the first engine and neglect the second, then panic when storms are quiet. Running both off one dataset smooths the year out.

The Short Version

Canvassing software that just maps territories and counts doors organizes your effort without improving your outcomes. The leverage is in targeting, knowing house by house which roofs are aging out and which ones a storm actually wore down, so your best people spend their hours at the doors most likely to become jobs. Express roof age as a range, treat storm exposure as odds, and let the physical inspection be the thing that decides what's real. Pair that targeting with disciplined territories, smart routes, honest dispositions, a clean field workflow, and tight compliance, and you turn a tired numbers game into a sharp, repeatable operation. Tools that rank which roofs are due, including RoofPredict, exist to get your closer to the right ladder. The rest, the honest inspection, the straight estimate, the homeowner's decision, is still yours to earn. That's exactly how it should be.

FAQ

What is roofing canvassing software, and how is it different from a CRM?

Canvassing software manages door-to-door field sales: it draws territories, assigns reps, tracks every door's status (knocked, not home, inspection set, signed), and reports activity. A CRM manages contacts and deals after a lead exists. The two integrate, the canvassing app feeds new leads into the CRM, but canvassing software is built around maps, routes, and field accountability rather than pipeline management. The best canvassing tools add a targeting layer that ranks which houses to knock first, which a generic CRM does not do.

How does software know which houses have roofs that need replacing?

It doesn't know with certainty, and any vendor claiming it does is overselling. Modern tools estimate roof age as a range using historical and current aerial imagery plus material and regional norms, and they model storm exposure for each individual roof based on hail size, wind, slope, and orientation. Combining an aging-out range with high per-roof storm odds tells you which houses are most likely due, so you inspect those first. The physical inspection is still what confirms actual condition.

Why express roof age as a range instead of an exact year?

Because an exact install date can't honestly be read from aerial imagery for a stranger's roof. What imagery can determine is roughly when the roof last visibly changed, plus condition cues, which yields a range like '14 to 19 years.' A range is enough to sort a street into likely-aging-out versus probably-fine, which is all targeting needs. Treat any single-year claim with suspicion; ranges are the honest and still-actionable output.

Is storm modeling proof that a roof is damaged?

No. Storm modeling gives you odds, an estimate of how likely a specific roof is to have suffered functional damage given the hail and wind it experienced. It is a reason to inspect, not evidence of damage and not a basis for any claim. Proof of damage comes only from a physical inspection where the roofer documents actual conditions. Treating odds as proof is the most common and most dangerous mistake in storm canvassing.

What metrics should I track to know if my canvassing is working?

Go beyond doors knocked. Track contact rate (answered/knocked), set rate (inspections/contacts), due rate (genuinely-due roofs/inspections), close rate (signed/due), cost per signed job, and sets-to-cash cycle time. Due rate is the direct scoreboard for targeting quality; close rate isolates your sales execution. Splitting those two apart tells you whether a weak week was a bad list or a bad sales process.

How big should a canvassing territory be?

Size territories by knockable doors, not acreage or street miles, because density varies enormously. A focused canvassing block in a tight subdivision covers far more doors than the same time on rural roads with long driveways. Keep territories contiguous and walkable, avoid splitting reps across highways or rivers, and re-cut territories around the actual storm swath after a major event rather than keeping old boundaries.

Do I need a permit to canvass door-to-door for roofing?

Often, yes. Many cities require a solicitation permit for door-to-door sales, and many neighborhoods maintain no-soliciting registries or run do-not-knock lists that carry legal weight. Requirements vary by municipality and sometimes tighten after major storms. Check the local clerk's office before a campaign and configure your software to exclude flagged addresses. Door-knocking itself isn't covered by federal Do-Not-Call rules, but phone and text follow-up is, so handle consent and opt-outs carefully.

How do I avoid looking like a storm-chasing contractor?

Be early, specific, and honest. Knock genuinely affected houses with a true, specific reason rather than blanketing a swath with generic 'free inspection' pitches. Document actual roof conditions, tell homeowners when a roof has life left, and never frame a storm model as proof of damage or promise claim outcomes. Targeting helps you read as the local pro who knocked the right houses, not the out-of-area crew working the whole ZIP code.

Where does RoofPredict fit in a canvassing operation?

RoofPredict provides the targeting layer: it ranks the houses inside a territory by which roofs are most likely due, combining a per-address roof-age range from aerial imagery with storm exposure modeled per individual roof. That lets a crew leader hand the highest-odds doors to the strongest closer and route the rest efficiently. It does not confirm damage, give exact install dates, or replace the inspection; the roofer still documents conditions and writes the estimate, and coverage stays between the homeowner and their insurer.

Will canvassing software pay for itself?

For most crews, yes, if the tool improves targeting rather than just tracking activity. The math is driven by due rate: moving from an untargeted neighborhood (maybe 1 in 4 inspections genuinely due) to a well-targeted one (more than half due) can double or triple signed jobs from the same labor. A single extra job typically covers the monthly software cost many times over, so the question is whether the tool actually changes which doors get knocked rather than only changing how they're logged.