To buy roofing leads by ZIP code, choose ZIPs on four tests — enough owner-occupied single-family homes, housing old enough that roofs are coming due, recent hail or wind on record, and a drive your crews can make — then filter inside each ZIP; per the Census Bureau’s 2020–2024 American Community Survey, the median US ZIP area holds 1,262 housing units, while half of all homes sit in ZIP areas of 12,535 or more.
A roofing company rarely buys a whole metro. It buys the ZIPs its crews can reach, usually from a list someone made years ago: where the first jobs came from, where the owner lives. The ZIP choice decides more about a lead program than the vendor does. Two ZIPs of about the same size can differ more than threefold in owner-occupied houses, as the Census table below shows, and a vendor will happily sell you either.
This page is for the owner or sales manager choosing those ZIPs and buying a roof list inside them. If what you need is the contact side — which seller, which phone, which ZIP the cut is made on — that is in homeowner lists by ZIP code.
Disclosure: Scout Data is our product and sells roofing lists by territory. Every housing figure below comes from the US Census Bureau, not from us.
What a ZIP actually holds
Start with the country. The American Community Survey 2020–2024 5-year estimates count 143.8 million housing units in the US, 65.2% of occupied units owner-occupied, and a median year built of 1980. Those are the averages every ZIP deviates from.
Then the ZIPs themselves. The Census publishes ZIP figures for ZIP Code Tabulation Areas, its approximation of the Postal Service’s delivery areas; post-office-box-only ZIPs have none. Across the roughly 33,600 of them in the 50 states and DC, in the same 2020–2024 estimates:
- The median ZIP area holds 1,262 housing units; a quarter hold fewer than 320 and a quarter more than 5,843.
- Nearly half of ZIP areas (45.5%) have fewer than 1,000 housing units, and together they hold only 3.7% of the country’s homes.
- Half of all US housing units sit in ZIP areas with 12,535 or more units.
So “a ZIP” is not a unit of supply. A rural ZIP will not fill a rep’s week; a suburban one can hold a season. Size every ZIP before you price it, and do it with the Census figure, not the vendor’s.
Three ZIPs, three different territories
Here are three suburban ZIPs of similar size, from the same Census release. The columns are the ones that matter for a roof list.
| ZIP | Housing units | Owner-occupied share | Owner-occupied detached homes | Median year built | Owner-occupied, built 2010 or later |
|---|---|---|---|---|---|
| 80015 (Aurora, CO) | 25,904 | 82.2% | 18,591 | 1992 | 2,276 |
| 75070 (McKinney, TX) | 25,707 | 52.7% | 11,695 | 2010 | 4,771 |
| 75243 (Dallas, TX) | 31,279 | 26.6% | 5,354 | 1982 | 215 |
Source: US Census Bureau, American Community Survey 2020–2024 5-year estimates for ZIP Code Tabulation Areas, read October 13, 2026. These are survey estimates with margins of error; housing units in 80015, for example, are ±743.
75243 is the largest of the three and the weakest roof territory: barely a quarter of its homes are owner-occupied, and it has fewer owner-occupied detached houses than either of the other two. A vendor quoting “31,000 homes” there is quoting apartments. 80015 is the opposite — four in five homes owner-occupied, most of them detached, and nearly two in three owner homes built in the 1980s or 1990s: a stock where roof age, not only storms, drives sales. 75070 sits between: plenty of owners, but more than a third of owner homes were built in 2010 or later, on roofs that are still young, beside a large band built in the 2000s. That ZIP splits by street, and the list has to split with it.
Four tests for a ZIP
| Test | Where to check, free | What rules a ZIP out |
|---|---|---|
| Owner-occupancy | ACS tenure and units-in-structure tables | Mostly renters or apartments, whatever the unit count |
| Housing age | ACS median year built and year-built bands | Most homes built in the last decade and on first roofs |
| Storm history | NOAA Storm Prediction Center storm reports | No hail or damaging wind near it in recent seasons |
| Distance | Drive time from your yard at 8 a.m. | An inspection and an install day you cannot staff |
Housing age is a floor, not a roof date. A 1985 house may be on its second or third roof, so the year built tells you a ZIP is in the market, not which houses are due; how roof age varies by state is in when do roofs need replacing, and how old the national stock is in American housing stock age.
Storm history is free to look up. NOAA’s Storm Prediction Center publishes local storm reports by date, with hail and wind reports back to 1955 in its SeverePlot archive; it labels the daily reports preliminary and points to NOAA’s Storm Data for the reviewed record. Scan the last few seasons around each ZIP. A ZIP hit twice in three years has fresh roofs on some streets and claim-ready roofs on others, which is exactly what a list has to separate — the field side of that is in storm damage leads.
Distance is the test most companies skip. Every lead in a far ZIP costs a longer drive for the inspection, the estimate, the install and every warranty call after it. Rank candidate ZIPs by drive time first and add outward from the yard, so the crews work adjacent ZIPs rather than islands.
The filters that matter inside a ZIP
Once the ZIPs are chosen, the list is what is left after the filters, and each one should take out homes you cannot sell, not homes you would rather not think about.
| Filter | Why it matters for a roof |
|---|---|
| Owner-occupied | A tenant cannot sign for a roof or file the claim |
| Single-family detached | Condo and many townhome roofs belong to the association, which buys roofs by bid, not at the door |
| Roof age past the local replacement point | The difference between a pitch and a courtesy inspection |
| Already re-roofed taken out | The fastest way a list wastes a rep’s day |
| Storm date on the row | Tells the rep which streets to work while the window is open |
| Roof material | Tile and metal run on a different clock and a different crew |
| Absentee owner | Sell by phone or mail, not at the door — a separate list |
Run the filters in that order and watch the count after each. In 80015 the Census already puts owner-occupied detached homes at 18,591 of 25,904 units; in 75243 the same two filters leave 5,354 of 31,279. The roof filters come after that, and they should shrink the list again, sharply. A list that barely shrinks under a roof-age filter is not reading roofs. How “already re-roofed” goes wrong when a vendor dates roofs only from permits is in why roofing permits miss re-roofs.
Test the vendor on one ZIP before you buy the territory
- Pick a ZIP you know. One where your crews have worked for years, so you can judge the list street by street.
- Ask for the count after each filter. Total homes, owner-occupied, detached, then each roof filter. Put the first three next to the Census figures for that ZIP. Allow for the survey’s margin and the years since, but a vendor showing far more owner-occupied detached homes than the Census is counting rentals, condos or stale records.
- Use your job history as the answer key. Every roof your company replaced in that ZIP should be missing from a due list, or carry your date. Every one that shows up as due is a row you would have paid for twice.
- Work the sample to the end. Knock or dial every row, log the outcomes, and compare inspections booked against the list you use now. The full method is in how to test a lead vendor with a free sample.
- Then add ZIPs outward. Adjacent ZIPs first, each one checked against the Census before it goes on the order.
Where Scout Data fits
Scout Data builds lists of homes by roof age and roof condition in your territory, with homes that were already re-roofed taken out. Each home carries one live mobile, matched by name to the owner on title and DNC-scrubbed before delivery. Start with one ZIP: ask for a free sample from the tens of thousands of roofs we have already read.
Frequently asked questions
How many homes are in a ZIP code?
Fewer than most buyers assume, and the range is huge. In the Census Bureau’s American Community Survey 2020–2024 5-year estimates, the median ZIP Code Tabulation Area in the 50 states and DC holds 1,262 housing units, a quarter hold under 320 and a quarter over 5,843. But half of all US housing units sit in ZIP areas with 12,535 or more, so the suburban ZIPs a roofer usually works are far larger than the median.
How do I choose ZIP codes for roofing leads?
Run four tests on each candidate ZIP: enough owner-occupied single-family homes to keep a crew busy, a housing stock old enough that roofs are coming due rather than in their first decade, hail or wind on record in recent seasons, and a drive your crews can make for an inspection and the install. The Census Bureau’s American Community Survey answers the first two for free; NOAA’s storm reports answer the third.
What filters should a roofing lead list by ZIP code use?
Owner-occupied, single-family detached, roof age past the local replacement point, homes already re-roofed taken out, and a storm date where there was one. Hold condos and townhomes with association-managed roofs, tile and metal roofs, and absentee owners as separate lists, because each needs a different pitch.
How do I test a roofing lead vendor before buying a territory?
Buy or sample one ZIP you already know. Ask the vendor for its home count after each filter and compare it with the Census figures for that ZIP; then check the sample against the roofs your own crews have replaced there, which should be missing or carry your date. Work the sample to the end before you add the next ZIP.