Stop Paying for Homes That Weren’t Hit

Here is the quiet tax most roofing companies pay every single storm season, and most of them have never bothered to calculate it:

When a hail swath rips through a metro, the actual damage footprint typically covers 5–15% of the parcels inside the affected ZIP codes. The other 85–95% are homes the storm did not touch. But every major property-data vendor still sells you the entire ZIP. You pay for 100% of the parcels, you can only monetize 8%, and nobody ever calls out the math.

This is the single biggest leak in the roofing industry's lead-generation economics. And once you see it, you can't unsee it.

The 8% problem

Say a supercell hits a DFW suburb. The NWS issues a warning polygon covering ~30 square miles. Inside that polygon, maybe 8,000 residential parcels fall inside the confirmed hail swath — the core plus the inner ring, where damage is actually likely. The surrounding ZIP codes contain a combined 120,000 homes.

A typical broad-data vendor asks you to choose between:

• Buy the ZIPs: 120,000 records. Paying full price for 112,000 homes that were never hit.
• Buy a city: Even worse. 400,000+ records for a single metro.
• Buy the county: Worst of all. You are now paying to call homes 40 miles from the storm.

None of those options are the actual impact zone. And none of them save you from the real cost, which is not the data bill.

The real cost is the calls you waste

Roofing companies that do cold outreach on storm leads know the rhythm. A caller dials 100 homes. Ninety of them did not get hail. Every one of those conversations takes 60–120 seconds to end politely — “I'm sorry, we did not get any damage over here.” Multiplied across a week of calling, this drains more than money. It drains morale, dial pace, and sequencing.

Here is what that actually looks like on a spreadsheet:

Broad ZIP-code list · 50,000 records

• Data cost (at $0.20/record): $10,000
• Calls dialed over 2 weeks: ~15,000
• Contact rate (cold, mostly uninvolved homes): ~8%
• Actual conversations: ~1,200
• Conversations about real hail damage: ~120 (10% of contacts)
• Closed inspections from those: ~25
• Effective cost per closed inspection: $400

Hyper-targeted storm-zone list · 3,500 records

• Data cost (at $0.50/record): $1,750
• Calls dialed over 2 weeks: ~3,000
• Contact rate (warm, storm actually hit): ~22%
• Actual conversations: ~660
• Conversations about real hail damage: ~460 (70% of contacts)
• Closed inspections from those: ~90
• Effective cost per closed inspection: $19

Same two weeks. Same team. One list produces 25 inspections at $400 of data cost each. The other produces 90 inspections at $19 of data cost each. Data efficiency went up 20×. And we have not even started counting the labor cost of the 12,000 dials the broad list burned through.

Why this happens

The property-data industry was not built for roofing. It was built for mortgage originators, insurance underwriters, and real-estate investors — audiences that care about ZIP-code-level market analysis, not 30-square-mile hail swaths. The tools and schemas reflect that. You get a lot of fields you will never use, filtered by a geography that has nothing to do with where the hail fell.

The typical workflow at a broad data vendor:

1. Log in. Look at a map.
2. Draw a polygon, or type in a ZIP code, or select a city.
3. Download 50,000 rows of CSV.
4. Hope your team can separate signal from noise.

The hail swath is nowhere in that process. Neither is the NWS warning polygon. Neither is the MRMS MESH hail radar data. You are on your own to figure out which of those 50,000 homes were actually damaged.

What hyper-targeting actually means

Hyper-targeted storm-zone data is not a different database. Under the hood, it is the same enterprise-grade property records. The difference is the starting point:

• Start with the storm polygon — the actual NWS warning geometry or the MRMS MESH hail swath, not a rectangular ZIP boundary.
• Tier by severity zone — core of the storm, inner ring, mid ring, outer fringe. You choose which zones matter based on how aggressive you want to be on the chase.
• Filter to the parcels that matter — residential vs commercial, owner-occupied, recent roof permit, solar permit, single-family, skip-traced or raw.
• Pay for what you pulled — not a subscription, not a seat minimum, not a contract. You see the count and the price before we charge anything.

The output is smaller. Much smaller. A typical hyper-targeted pull for a significant storm event lands between 1,500 and 8,000 records, depending on the severity tiers you select. That feels tiny compared to a 50,000-row ZIP dump. It is also the reason your close rates go up when you start working it.

What to look for when picking a data vendor

If you are evaluating roofing-specific storm data (including us), these are the questions that actually matter:

1. Can you target by storm geometry, not ZIP code? If the answer is “draw your own polygon,” you are still doing the vendor's job for them.
2. Do you see the record count and cost before paying? Any platform that makes you commit before you know the number is a platform that benefits from your surprise.
3. Can you layer severity tiers? “Core only” vs “core + inner” should be a checkbox, not a support ticket.
4. Is there a contract? Storm work is by nature lumpy. If you are forced to pay monthly when nothing happened, the economics will not survive an off season.
5. How fast from a map to a CSV? Two minutes is the new industry standard. Anything longer means a chaser from three states away is already working your neighborhood.

The bottom line

Every dollar spent on a home that was not hit is a dollar wasted. Every minute your caller spends apologizing to an unaffected homeowner is a minute they could have spent on a house that actually has damage. And every week your team slogs through a 50,000-row ZIP dump is a week a better-tooled competitor is closing jobs your list will never reach.

The math on broad-data targeting was already ugly five years ago. With severity rising and claim economics tightening, the math in 2026 is going to get uglier still. The roofers who see that early and switch to hyper-targeted data are the ones who will spend less and close more — every week, every storm, every season.

That is the whole thesis behind Roofers Radar. If you want to see what a real storm-zone pull looks like for your market, spend two minutes on the product. Upload any hail map and we will show you the count and the cost before you commit.