Data Sources for Roofing Territory Targeting Compared

Short Answer

The best data sources for roofing territory targeting fall into five buckets, and no single one wins on its own. Use public housing data (the Census Bureau's American Community Survey for age-of-housing-stock and ownership, and the Building Permits Survey for new construction and reroof activity) to understand where old roofs cluster. Use storm and hazard data (the NOAA Storm Events Database, the Storm Prediction Center, and the National Weather Service hail and wind records) to find where roofs were recently damaged. Use parcel and assessor data (county tax records) for lot-level facts like year built, lot size, and owner-occupancy. Use mail-delivery data (USPS Every Door Direct Mail route counts) to size and price a saturation drop. And use roof imagery and derived signals (aerial and satellite imagery, condition models) to refine the list a property at a time.

Public sources are free, authoritative, and great for strategy — deciding which ZIP codes and neighborhoods to work — but they are coarse: the ACS reports neighborhood averages, not individual addresses. Parcel and imagery data are granular enough for list building — picking the actual homes to mail — but they cost money, vary in quality county to county, and need cleaning. The practical answer for most roofing companies is a layered stack: start with free public data to choose the territory, then add a paid parcel list (or a platform that has already joined these layers) to build the mailing list, then validate the most expensive drops against storm and imagery signals.

This guide compares each source on coverage, granularity, freshness, cost, and what it can and cannot tell you, with worked examples, a scoring worksheet, and a buyer's checklist. The short version: free government data is the foundation every roofer should build on, paid parcel data is the workhorse for list building, storm data is the trigger, and imagery is the finishing layer. Match the source to the decision you are making, and never mistake a neighborhood average for an individual roof.

Sources checked: June 20, 2026.

Why Data Sources for Roofing Territory Targeting Are the Whole Game

Roofing is a local business with a brutal acquisition-cost problem. You cannot knock every door, mail every address, or buy every click, so the question that decides your margin is which homes do I spend on? Good data narrows a 40,000-home market down to the 6,000 homes most likely to need a roof in the next two years, and that narrowing is the difference between a campaign that returns three dollars per dollar and one that loses money.

Most owners think about data the moment they are about to buy a list. That is too late. The data decision actually happens earlier, when you choose a service area and a strategy. A territory full of 1990s-and-older housing in a hail-prone county is a different business than a territory of five-year-old subdivisions, and the public data to tell those two apart is free and sitting on a government website. Skip that step and you can run a flawless mail campaign into a market that has no demand yet.

The honest framing is that data sources are tools, and each tool answers a different question. The ACS answers "where is the old housing?" The NOAA Storm Events Database answers "where did roofs just get hit?" The county assessor answers "what year was this house built and who owns it?" USPS route data answers "what will it cost to blanket this carrier route?" Imagery answers "does this specific roof look aged or replaced?" Buy or use them as if they are interchangeable and you will be disappointed; layer them deliberately and you build a targeting engine.

The Five Categories of Roofing Territory Data

Before comparing individual sources, it helps to see the landscape. Almost every data source a roofing company will touch belongs to one of five categories, and each category has a characteristic strength and a characteristic weakness.

The pattern to notice: as you move down the table, the data gets more granular and more expensive, and it moves from strategy (which market) to execution (which house). The free public sources at the top are where every roofer should start, and the paid sources at the bottom are where you spend money only after the free data has told you the market is worth working.

A second pattern: freshness runs the opposite direction from cost in places. Storm data is free and fresh (updated within days). Census housing data is free but lags a year or more. Parcel data is paid and can be stale (assessors update on their own cycles). Imagery freshness depends entirely on when the provider last flew or imaged your area. Never assume "paid" means "current."

Public Housing Data: The Free Foundation

The single most underused resource in roofing marketing is the U.S. Census Bureau. It is free, it is authoritative, and it answers the most important strategic question for a roofer: where is the old housing stock?

American Community Survey (ACS)

The American Community Survey is an ongoing survey that publishes detailed housing characteristics down to the census-tract and block-group level. For roofers, the relevant tables are year structure built, tenure (owner- vs renter-occupied), median home value, and median household income. Because roof age correlates loosely with home age, the "year structure built" distribution is a strong first proxy for where roofs are aging out.

What it does well: it covers the entire country, it is free, and it lets you rank neighborhoods by the share of homes built before, say, 2005. A tract where 70% of homes were built in the 1980s and 1990s is far more likely to have roofs in or near the replacement window than a tract that is mostly post-2015 construction.

What it does not do: the ACS reports aggregates, not addresses. It tells you a tract is "mostly 1990s housing," not which specific houses those are, and certainly not which roofs have already been replaced. It also carries a sampling lag — the data describes the recent past, not this week. Treat the ACS as a way to choose and rank territories, never as a mailing list.

Building Permits Survey

The Census Building Permits Survey tracks residential construction permits by place and county. For roofers it has two uses. First, high new-construction counts flag young housing stock you probably want to avoid for replacement work (new roofs, but maybe future maintenance or builder relationships). Second, in jurisdictions that require reroof permits, permit volume hints at how much replacement activity is already happening — a competitive-intensity signal.

HUD and other federal housing data

The Department of Housing and Urban Development publishes housing datasets that can help with demographic and affordability context, useful when you are deciding whether a territory's homeowners can self-fund a reroof or will lean heavily on insurance. It is supporting data, not primary targeting data, but it rounds out the strategic picture.

How to actually use public housing data

Here is the practical workflow. Pull "year structure built" for every tract in your metro, calculate the percentage of homes built in your target window (commonly 15–30 years ago, depending on the dominant roofing material), and rank tracts. Overlay owner-occupancy, because renters rarely buy roofs. The result is a heat map of strategic priority — the neighborhoods worth investing in — before you have spent a dollar on a list.

PUBLIC-DATA TERRITORY SCREEN (free, ~2 hours)
1. Census ACS: pull "Year structure built" by tract for your metro.
2. Compute % of housing units in your target age band per tract.
3. Pull "Tenure" — keep tracts above ~60% owner-occupied.
4. Pull "Median home value" — flag tracts that can self-fund or insure.
5. Building Permits Survey: note high-new-construction places to deprioritize.
6. Rank tracts by (age-band % x owner-occ %); pick your top 8–12.
RESULT: a prioritized territory map, before buying any list.

Storm and Hazard Data: The Trigger Layer

For most roofing companies, weather is the largest single driver of replacement demand. A hail or high-wind event can create more qualified leads in one afternoon than a year of age-based marketing. The good news: the authoritative storm data is free and remarkably granular by category.

NOAA Storm Events Database

The NOAA Storm Events Database is the archived, official record of significant weather events, including hail (with reported diameters) and thunderstorm wind (with estimated speeds). For a roofer it is the source of truth for "did a damaging event actually happen here, and how big was it?" Hail under three-quarters of an inch rarely damages modern shingles; golf-ball size (1.75") and up typically produces widespread, claimable damage. The database lets you confirm both the location and the severity before you spend on a campaign.

Storm Prediction Center and the National Weather Service

The Storm Prediction Center publishes daily storm reports and outlooks — useful for near-real-time response, before the archived database catches up. The National Weather Service issues warnings and post-event summaries, and its hail safety and information pages help you interpret what a given hail size means for roofs. Together these give you a fast layer (SPC reports, day-of) and a definitive layer (Storm Events Database, archived) for the same events.

FEMA disaster data

After major declared disasters, FEMA data identifies federally declared areas, which can indicate concentrated damage and an active rebuilding environment. This is coarse (county-level declarations) but useful for understanding the macro environment you are working in.

What storm data can and cannot do

Storm data tells you where damaging weather occurred. It does not tell you which individual roofs were damaged — that requires inspection, and no dataset substitutes for a licensed roofer on the property. The honest use is to define a storm swath (the footprint of confirmed damaging hail or wind) and then target homes inside that swath, while being clear with homeowners that a public weather record is a reason to inspect, not proof their roof is damaged.

Parcel and Assessor Data: The List-Building Workhorse

When it is time to build an actual mailing list — specific addresses, not neighborhood averages — parcel data is the workhorse. County tax assessors maintain records on every parcel: year built, square footage, lot size, owner name, owner-occupancy (via mailing-address mismatch), and assessed value. This is the only widely available source that gives you year built at the individual address, which is the closest free-ish proxy for roof age.

Where parcel data comes from

Assessor data originates with each county. Some counties publish it free on a GIS portal; most roofers, though, buy it cleaned and nationally standardized from a data vendor or get it bundled inside a marketing or roofing platform. The raw county data is messy — every county uses different field names, formats, and update cadences — so the value a vendor adds is standardization and joining, not the underlying facts.

Strengths

Parcel data is granular (address-level), broadly available (every property is on a tax roll), and rich (year built, lot, owner, value in one record). For age-based targeting it is the backbone: you filter to owner-occupied homes built in your target window and you have a defensible list.

Weaknesses to plan for

Three things go wrong with parcel data, and you should plan for all three:

1. Staleness. Assessors update on their own cycles; a record may reflect last year's owner or an out-of-date characteristic. Owner-occupancy flags are derived and imperfect.
2. Year-built is not roof age. A 1995 house may have a 2018 roof (already replaced) or an original 1995 roof (overdue). Year built is a probability signal, not a roof-age fact — a point worth repeating because it is the single most common over-claim in roofing marketing.
3. County-by-county quality variance. Coverage and field completeness differ wildly. A metro that spans five counties may have five different data qualities, and your list inherits the worst of them.

Buying parcel data: a comparison frame

The DIY route is genuinely free if you have someone who can wrangle GIS exports, but it does not scale across counties. A list vendor scales but charges per record and leaves you to join storm and imagery layers yourself. A platform that has already joined parcel, storm, and imagery data trades a subscription for not having to do that integration — the right call once you are running campaigns repeatedly rather than one-off.

Mail-Delivery Data: Sizing and Pricing the Drop

Once you know which homes you want to reach, you need data to size and price the physical mail. This is where USPS data earns its place.

USPS Every Door Direct Mail route data

Every Door Direct Mail lets you select carrier routes on a map and see the number of residential (and business) addresses on each route, along with demographic estimates. For saturation strategies — blanketing every home in a storm swath or an aging neighborhood — EDDM route counts let you size the drop precisely and estimate postage before you commit. The broader USPS business mail resources cover the options for targeted (addressed) mail when you want to hit a specific list rather than whole routes.

Why deliverability data matters for targeting

Mail-delivery data is not only about postage. Address standardization and deliverability checks (running your list against USPS address validation) remove undeliverable and vacant addresses before you pay to print and mail them. A list that looks 6,000 strong but contains 400 undeliverable addresses is a list you are about to waste money on. Deliverability is a targeting decision disguised as a logistics one.

The cost model you must get right

Here is the honest mail-cost framing, because it is where roofers most often miscalculate. The report or scoring layer (deciding which homes to target) is one cost; the physical mailers are a separate, real, dollar-denominated cost billed per piece, with volume discounts as send size grows. A common list price is around $0.68 per piece, with discounts kicking in at higher volumes (for example, meaningful percentage reductions at 1,000, 2,500, and 5,000+ pieces). Always compute an honest dollar total for the physical mail — never assume targeting and printing are the same line item.

MAIL DROP SIZING WORKSHEET
A. Target homes after filtering (parcel + storm + dedupe) = _____
B. Undeliverable/vacant removed (USPS validation)        = _____
C. Net mailable count (A - B)                            = _____
D. Per-piece print+mail cost (with volume discount)      = $____
E. Drops planned (touches per home)                      = _____
F. Total mail spend = C x D x E                          = $____
G. Conservative response rate (e.g., 0.5%)               = ____%
H. Expected jobs = C x G                                 = _____
I. Cost per job = F / H                                  = $____
CHECK: Is I comfortably below your average job margin? Y / N

Imagery and Derived Signals: The Finishing Layer

The newest and most granular category is roof imagery and the condition signals derived from it. Aerial and satellite imagery, combined with models that estimate roof material, footprint, and apparent condition, can refine a list one roof at a time.

What imagery adds

Year-built tells you a 1998 house probably has an older roof. Imagery can suggest whether this specific roof looks weathered, recently replaced, or a different material than expected. That extra signal helps you avoid mailing a home that visibly got a new roof last year, and helps you prioritize roofs that look aged. On your most expensive drops — the late, narrowed, high-cost touches — this refinement can lift response meaningfully.

The honest limits of imagery

Imagery is a signal, not an inspection. A model looking at a top-down photo cannot certify a roof's condition, age, or remaining life — only a licensed roofer on the property can. Imagery freshness varies by provider and area; a "current" image may be a year or two old. And imagery-derived condition scores are probabilistic. Use them to prioritize, never to tell a homeowner their roof is damaged from a photo. That distinction is both ethical and, frankly, what keeps you out of trouble.

Cost and access

Imagery and derived signals are the most expensive category and usually come through a paid provider or a platform. For a small operator running occasional campaigns, imagery may be overkill; for a company mailing tens of thousands of pieces a quarter, the refinement can pay for itself by trimming waste on the priciest drops.

Putting It Together: A Layered Targeting Stack

No single source is the answer. The winning approach layers them so each does the job it is best at. Here is the canonical stack, from strategy to execution:

Read top to bottom, the stack moves from free-and-coarse to paid-and-granular, and from "which market" to "which house." Most roofers should always do layers 1, 2, and 4 (they are free or cheap and high-leverage), should do layer 3 as soon as they are building real lists, and should add layer 5 once volume justifies it.

A worked example (illustrative)

Imagine a contractor — call them a mid-size company working a single metro — deciding where to spend next quarter's mail budget. They start with the ACS, rank tracts by share of 1995–2010 housing, and keep the top ten owner-occupied tracts (layer 1). They check the NOAA Storm Events Database and find a confirmed 1.75" hail swath cutting through three of those tracts last month (layer 2). They build a parcel list of owner-occupied homes built before 2010 inside the swath (layer 3), run it through USPS validation, and use EDDM counts to confirm the saturation cost on the surrounding routes (layer 4). For the final, narrowed third drop, they remove homes whose imagery suggests a recent reroof (layer 5). Every layer used a source for exactly what it does best. This is illustrative — your tracts, swath, and counts will differ — but the sequence is the repeatable part.

Cost Comparison: What Each Source Actually Costs

Roofers want the bottom line on cost. Here it is as ranges, because real pricing varies by vendor, volume, and geography. Treat these as relationships, not quotes.

Two cost truths worth internalizing. First, the most valuable sources (Census and NOAA) are free — your competitive edge is using them, not buying them. Second, the line item that actually moves your budget is the physical mail, billed per piece in real dollars with volume discounts, not the targeting data itself. Spend your analysis on getting the list right so the dollars you spend on postage land on the right homes.

Coverage and Granularity Compared

The two axes that decide whether a source fits a job are coverage (how much of the country it spans) and granularity (how fine — tract, route, or address). Here is the comparison on those axes.

The takeaway: only parcel data and imagery are truly address-level. Everything else is an aggregate or a footprint. That is exactly why you layer — you use the aggregates to choose the area and the address-level sources to choose the homes.

Freshness Compared: Don't Mail on Stale Data

Freshness is the most overlooked dimension. A perfect list built on a stale source is still a bad list.

The defensive move is to date-stamp every source in your process and re-pull anything time-sensitive. Storm data must be current to the event. USPS validation should run right before each drop. Parcel data should be refreshed at least annually. The ACS is fine to refresh yearly because it changes slowly.

Common Mistakes With Roofing Territory Data

The same handful of mistakes cost roofers money over and over. Avoid these.

• Treating year built as roof age. It is a probability signal, not a fact. A confident "your roof is 25 years old" based on a tax record is wrong as often as it is right — and it erodes trust when the homeowner replaced the roof in 2019.
• Treating a neighborhood average as a list. The ACS says a tract is "mostly old housing"; it does not name the houses. Mailing every address in an "old" tract wastes money on the new and already-reroofed homes inside it.
• Mailing a storm swath without confirming the event. Always verify hail size and location in the NOAA record. A small-hail event under three-quarters of an inch rarely justifies a damage campaign, and over-claiming damage from a marginal storm invites complaints.
• Skipping deliverability. Paying to print and mail undeliverable or vacant addresses is pure waste, and it is entirely preventable with USPS address validation.
• Buying granular data before validating the market. Spending on parcel and imagery in a territory the free Census data would have told you to skip.
• Ignoring county-quality variance. Assuming a multi-county list is uniformly good. It usually is not; the worst county drags the list down.
• Over-claiming from imagery. A top-down photo is a prioritization signal, not an inspection. Telling a homeowner their roof is damaged from a photo is both inaccurate and a credibility risk.

Compliance and Honesty in How You Use Data

Targeting data touches marketing rules and consumer trust, so a few guardrails belong in every roofer's process.

When you advertise, the FTC's advertising and marketing guidance applies: claims must be truthful and substantiated. "Your neighborhood was hit by hail — schedule a free inspection" is fine if a storm record supports it; "your roof is damaged" based on a dataset is not, because no dataset inspects a roof. For any phone follow-up, the Telemarketing Sales Rule and Do Not Call rules govern who you may call. If you use email, the CAN-SPAM rules apply. And because storm-chasing scams have given the trade a bad name, the FTC's guidance on avoiding home-improvement scams is worth reading just to see how not to come across.

The honest line to hold: data tells you where to offer an inspection, not what condition a roof is in. Keep that distinction visible in your copy and you stay both compliant and credible.

A Buyer's Checklist for Evaluating a Data Source

When a vendor or platform pitches you "roofing data," run it through this checklist before you pay.

DATA SOURCE EVALUATION CHECKLIST
[ ] Coverage: Does it cover ALL counties in my service area?
[ ] Granularity: Address-level, route-level, or tract-level? (Match to the job.)
[ ] Freshness: How old is the data, and how often is it refreshed?
[ ] Source of truth: Where does it originate (assessor? NOAA? imagery flyover?)?
[ ] Year-built accuracy: Is it parcel-sourced? How complete by county?
[ ] Owner-occupancy: How is it derived, and how is it validated?
[ ] Deliverability: Is USPS address validation included or extra?
[ ] Storm join: Does it incorporate NOAA storm data, and how current?
[ ] Imagery: If condition is claimed, is it labeled a signal (not an inspection)?
[ ] Cost shape: Per-record, subscription, or bundled? What's the mail cost?
[ ] Export: Can I get my list out (CSV) for printing/mailing?
[ ] Claims hygiene: Does it avoid telling me a roof IS damaged from data alone?

If a source can't answer the freshness, coverage, and granularity questions clearly, treat its other claims with suspicion. And if a source claims to know a roof's condition or age as fact from data alone, that is a red flag — those are inspection determinations, not data outputs.

Regional and Seasonal Variation

Data strategy is not one-size-fits-all. It shifts by region and season.

Hail-alley markets (the southern and central plains) lean heaviest on storm data — the NOAA Storm Events Database is the primary trigger, and age data plays a supporting role. Campaigns are reactive and fast, and the swath is the territory after a major event.

Coastal and high-wind markets weight wind events and, where relevant, IBHS FORTIFIED considerations into the picture, since resilient-roofing programs change both demand and messaging. Storm data here is more about wind speed than hail size.

Stable-weather markets (much of the interior West and parts of the Northeast) lean on age data because storm-driven demand is rarer. Here the Census ACS and parcel year-built are the primary engine, and campaigns are proactive and steady rather than reactive.

Seasonally, storm sources spike in relevance through spring and summer convective season, while age-based targeting runs year-round. Permit data has its own rhythm — construction activity rises in warm months — which matters if you use permits to read competitive intensity. The practical rule: weight your stack toward storm data in storm season and storm-prone markets, and toward age and parcel data in calm seasons and calm markets.

Decision Framework: Which Source for Which Job

When you are unsure which source to reach for, match the decision to the source.

The meta-rule: free public data first (it answers the biggest, cheapest-to-answer questions), paid data only where it adds address-level precision the free data cannot.

Where RoofPredict Fits

You can do everything above by hand for one neighborhood: pull the ACS tables, check the NOAA Storm Events Database, download a county's parcel file, clean it, run USPS validation, and eyeball imagery. It works — for one neighborhood, once. The problem is repeatability and scale across a whole territory and a full mail calendar.

RoofPredict is the operational layer that joins these data sources for you. It scores which properties in a territory are most likely to need roof work — using property age and characteristics, storm and hail exposure history, and roof-imagery signals — and turns that scored list into targeted direct-mail campaigns and professional roof reports. In other words, it does the layering described in this guide (strategy → trigger → list → hygiene → refine) across an entire service area, repeatably, instead of one hand-built list at a time.

On cost, the honest model: a subscription with credits covers the roof reports — one report per home, no matter how many times you mail that home; credits are not consumed per mailer. The mailers themselves are billed in real dollars, per piece (a common list price is around $0.68/piece), with volume discounts as send size grows (for example, meaningful percentage reductions at 1,000, 2,500, and 5,000+ pieces). Nothing is charged for a mail run until the proof is approved and the mailers go to print. Lead with the count if you like, but always look at the honest dollar total for the mail.

Guardrail: RoofPredict's score is a prioritization signal built from data — it tells you which homes are most worth an inspection offer. It does not inspect, climb, or certify a roof, it does not prove a roof's age or that a storm caused damage, and it does not decide or guarantee any insurance claim. Those determinations belong to a licensed roofer, a public adjuster, and the building department — not to any dataset or software. Use the data to decide where to spend; let the professionals on the property decide what the roof needs.

For Roofers: A 30-Day Plan to Build Your Data Stack

If you are starting from scratch, here is a concrete month to assemble a real targeting stack without overspending.

Week 1 — Strategy (free). Pull Census ACS "year structure built" and "tenure" for every tract in your metro. Rank tracts and pick your top eight to twelve. Glance at the Building Permits Survey to deprioritize heavy-new-construction areas. Cost: your time.

Week 2 — Trigger (free). Set up a habit of checking the NOAA Storm Events Database and SPC reports. Map any confirmed damaging hail or wind swaths over your priority tracts. Decide whether your near-term play is storm-reactive or age-proactive based on what you find.

Week 3 — List (paid, small). For your top tracts (or a confirmed swath), acquire parcel data — DIY from county portals if you have one county and a technical hand, or a list vendor / platform if you span several. Filter to owner-occupied homes in your target age band. Run USPS address validation.

Week 4 — Size, refine, send. Use EDDM counts to confirm saturation cost or finalize your targeted list, compute the honest dollar total with the mail-cost worksheet, optionally trim with imagery on your most expensive drop, and launch a test drop. Measure response, then iterate.

Run that loop and you will have, in a month, the layered stack most roofers never build: free strategy data, free trigger data, paid address-level list data, deliverability hygiene, and optional imagery refinement — each used for exactly what it does best.

Key Takeaways

• No single source wins. Roofing territory targeting is a layered stack: public housing data for strategy, storm data for triggers, parcel data for lists, USPS data for sizing, imagery for refinement.
• Free government data is your foundation. The Census ACS (age, ownership) and NOAA Storm Events Database (confirmed events and severity) are free, authoritative, and high-leverage. Using them is the edge, not buying them.
• Only parcel data and imagery are address-level. Everything else is an aggregate or a footprint. Use aggregates to choose the area, address-level sources to choose the homes.
• Year built is a probability signal, not roof age. The most common and costly over-claim in roofing marketing. Never tell a homeowner their roof's age or condition from a dataset.
• Freshness matters as much as accuracy. Re-pull time-sensitive sources: storm data per event, USPS validation per drop, parcel data at least yearly.
• Mind the cost shape. Targeting data is mostly free or cheap; the physical mail is the real dollar cost, billed per piece with volume discounts. Spend your analysis getting the list right.
• Stay honest and compliant. Data tells you where to offer an inspection, not what a roof's condition is. Keep that line visible to stay both within FTC guidance and credible with homeowners.

FAQ

What is the best data source for roofing territory targeting?

There is no single best source — the right answer is a layered stack. Use the Census American Community Survey for strategy (where old, owner-occupied housing clusters), the NOAA Storm Events Database for storm triggers, county parcel/assessor data for address-level list building, USPS data for sizing and deliverability, and imagery for per-roof refinement. The "best" source depends on which decision you are making: strategy, trigger, list, or refinement.

Where can roofers get property data for territory targeting for free?

The two most valuable free sources are the U.S. Census Bureau (the American Community Survey for housing age and ownership by tract, plus the Building Permits Survey) and NOAA (the Storm Events Database, Storm Prediction Center, and National Weather Service for storm and hail records). USPS also lets you view Every Door Direct Mail route counts for free. These cover strategy and trigger decisions at no cost; address-level list data usually requires paid parcel data.

How accurate is year-built data for estimating roof age?

Year built is a useful probability signal, not a fact about roof age. A home built in 1998 may have an original roof that is overdue or a roof replaced in 2019 — the tax record cannot tell which. Use year built to prioritize neighborhoods and build target lists, but never tell a homeowner their roof's actual age or condition based on a dataset; only a roof inspection can determine that.

What is the difference between parcel data and Census housing data?

Census data (the ACS) is aggregated to neighborhoods — it tells you a tract is "mostly 1990s housing" but does not name the houses. Parcel/assessor data is address-level — it gives year built, owner, and characteristics for each individual property. Use Census data to choose which areas to work and parcel data to build the actual mailing list of specific homes within those areas.

How do I find out if a storm actually hit a neighborhood?

Check the NOAA Storm Events Database for the archived, official record with reported hail diameters and wind speeds, the Storm Prediction Center's daily storm reports for fast day-of confirmation, and the National Weather Service warnings and post-event summaries. Confirm both the location and the severity — hail under three-quarters of an inch rarely damages modern shingles, while golf-ball-sized (1.75") and larger usually produces widespread, claimable damage.

Do I need to buy data, or can I run a campaign on free sources alone?

You can choose a strong territory and confirm storm triggers entirely on free Census and NOAA data. To build an actual address-level mailing list, though, most roofers need paid parcel data (or a platform that includes it), because the free aggregates do not name individual homes. A sensible sequence is: validate the market with free data first, then spend on address-level data only where the market is proven.

What does roofing property data typically cost?

It varies by source and shape. Census and NOAA data are free. USPS EDDM route data is free to view, with postage billed on send. County parcel data ranges from free (DIY portal) to a per-record fee from a vendor. Imagery and condition signals are the most expensive category. The largest real cost is usually not the data at all — it is the physical mail, billed per piece in dollars with volume discounts at higher send sizes.

How fresh does my roofing territory data need to be?

It depends on the source. Storm data must be current to the event you are responding to. USPS address validation should run right before each drop to remove vacants and undeliverables. Parcel data should be refreshed at least annually because assessors update on their own cycles. Census ACS data changes slowly, so a yearly refresh is fine. Always date-stamp your sources and re-pull anything time-sensitive.

What is the difference between EDDM and a targeted mailing list?

Every Door Direct Mail (EDDM) lets you blanket entire carrier routes by selecting them on a map — it is the cheapest, fastest way to saturate an area, but you cannot exclude renters, vacants, or new roofs within a route. A targeted (addressed) list lets you mail specific homes filtered by age, ownership, and other criteria, which costs more per piece but cuts waste. Many roofers use EDDM for fast saturation drops and targeted lists for refined follow-ups.

Can roof imagery tell me which homes need a new roof?

Imagery and condition models are a prioritization signal, not an inspection. They can suggest a roof looks weathered or was recently replaced, which helps you trim and rank a list, but a top-down photo cannot certify a roof's condition, age, or remaining life. Use imagery to prioritize the homes worth offering an inspection — never to tell a homeowner their roof is damaged from a photo.

How do I target the right homes after a hail storm?

Define the storm swath from the NOAA Storm Events Database (confirmed location and hail size), then build a parcel list of owner-occupied homes inside that swath, run it through USPS address validation, and size the drop with EDDM counts for saturation or a targeted list for precision. On the most expensive follow-up drops, optionally remove homes whose imagery suggests a recent reroof. Each layer does one job — trigger, list, hygiene, refinement.

Is it legal to use storm and property data for roofing marketing?

Using public storm and property data to decide where to offer inspections is legitimate, but how you advertise is regulated. FTC advertising rules require claims to be truthful and substantiated, so "your area was hit by hail — schedule a free inspection" is fine with a storm record behind it, while "your roof is damaged" from data alone is not. Phone follow-up falls under Do Not Call and the Telemarketing Sales Rule, and email under CAN-SPAM. Keep the line that data tells you where to offer an inspection, not what condition a roof is in.

Why does parcel data quality vary so much by county?

Parcel data originates with each county assessor, and every county uses different field names, formats, update cycles, and levels of completeness. A metro that spans five counties can have five different data qualities, so a multi-county list inherits the weakest county's gaps. Vendors and platforms add value by standardizing and joining these county feeds, but the underlying variance is why you should always check coverage and completeness county by county.

Should a small roofing company invest in imagery data?

For a small operator running occasional campaigns, imagery is often overkill — the free Census and NOAA data plus a basic parcel list will carry you. Imagery earns its cost once you are mailing large volumes and want to trim waste on your most expensive, narrowed drops, where avoiding already-reroofed homes lifts response. Start with the free and cheap layers, prove the model works, and add imagery only when send volume justifies the spend.

How do I combine multiple data sources into one targeting list?

Layer them in sequence: use Census ACS to pick priority tracts, NOAA data to confirm any storm trigger, parcel data to pull address-level homes in your target age band, USPS validation to remove undeliverables, and optionally imagery to trim recently reroofed homes. You can do this by hand for one neighborhood, or use a platform that has already joined parcel, storm, and imagery layers so the stack is built and repeatable across your whole territory.