Your ads go live while the hail is still falling — or the wind is still taking shingles.
And — this is the part most storm tools get wrong — they stay off where nothing happened. A quarter-inch hail report is not a roof. Spending against it is how a storm budget disappears before the storm that mattered arrives.
What puts you live
1.25″ is the documented bar for asphalt shingle damage. Below it you are buying clicks from people whose roofs are fine. Wind needs no such bar — a measured gust or a report that says the roof came off is already the answer.
Can I trust where this comes from?
It is the weather service, quoted verbatim — including the misses.
Any tool can show you what it caught. The list beside this shows what it ignored on the same night, which is the harder half and the one that decides whether your budget survives to the storm that matters.
Two hail reports crossed the threshold and put ZIP codes live. Two did not, and nothing was spent against them. The fifth is not hail at all — a wind report naming the roof it took off, which is as direct a signal as this feed produces.
The refusals are the product. A system that fires on everything is a system that has spent your season by May.
Ground reports, 12 September
Source: NOAA / NWS Local Storm Reports, 2026-09-12. Reproduced exactly as filed. In that one week nationally there were 47 reports naming roof, shingle or siding damage — and not one of them came from hail.
You are probably already paying for one of these
After the storm, who does the next thing?
| Tool | What it does when a storm hits | Who acts next | Yours alone? |
|---|---|---|---|
| GAF WeatherHub | Emails you when a roof you installed and warrantied sat under 60 mph wind or 1″ hail. | You — call your past customers | Yes, but only your own past customers |
| HailTrace | Street-level hail footprint from real meteorologists, with date-of-loss reports. | You — work the map, knock the doors | No. The same map is sold to every contractor in your market |
| FirstAlert | Alerts you to a 911 dispatch in under sixty seconds. | You — get in the truck and drive | Yes, for six months |
| WeatherAds | Gives you a tool to build weather triggers in. | You — configure it, run it, watch it | No. Self-serve, anyone can buy it |
| Muster. | Switches your Google Ads on within minutes of an NWS ground report. | The customer — they are searching, and they call you | Yes — one contractor per trade, in the ZIP codes you run |
Four of these end with you making the call. Muster ends with the homeowner making it.
Does this actually fit roofing?
Hail is where the money is. Wind is what fires in between.
Start with the part that pays you. Across severe convective storms, hail accounts for 50–80% of insured losses — that is Allianz Commercial's range, matched by Swiss Re, and it is why a hail season is a roofing company's year. Nothing here argues with that.
The problem is when it arrives. 82% of hail lands between March and July, and it is heavily inland — Texas, Kansas, Missouri, Nebraska and the Dakotas carry most of it. A campaign that only watches hail is a campaign that sits idle for two thirds of the year and does nothing at all in a lot of the country.
In the weather-service feed, hail is 29% of the reports that qualify a roofing campaign; measured wind and roof-damage reports are the other 71%, and they arrive in every month and every state. Both numbers are true and they measure different things — one hail report can put a whole town's roofs in play, where one wind report is often a single house. That gap is the point: hail is denser, wind is constant.
We measured what that is worth. We took six counties on the North Carolina coast, ran four years of weather-service reports through the same triggers we would set up for you, and asked one question: sign in August, and how long until your first activation? Watching hail alone: 264 to 282 days — every single year, never once inside the first six months. Counting measured wind and roof-damage reports as well pulls it to a median of 165 days. We picked that market on purpose because it is the hardest one — on the coast, hail is rarest. Your county almost certainly does better, and you can check yours in about twenty seconds. If your storms are mostly water rather than wind, restoration is the more honest conversation.
Our own analysis, not a vendor's. We ran four years of National Weather Service ground reports for six counties on the North Carolina coast through the same triggers we build for clients, starting the clock in August. Left: watching hail only. Right: counting measured wind and roof-damage reports too. We used a coastal market deliberately — it is where hail is rarest, so it is the worst case, not the best one.
Am I the right contractor for this?
Some of you should not buy this. Here is which.
This fits you if
- You work a hail market, or anywhere measured wind and roof-damage reports are what fire
- You run your own Google Ads account and know roughly what a click costs you
- You can put an inspector out inside 48 hours of a storm
- You are already competing on storm work and losing the first week of it to somebody faster
- You handle insurance claims work and know the value of documented, timestamped hail size
It does not fit you if
- Your hazard is mostly water rather than wind — flooding and mitigation work. Restoration is the better fit
- You are canvassing-led and do not run meaningful paid search. This amplifies an ad account; it does not replace a door-knocking crew
- Someone else already holds the ZIP codes you run. One roofer per trade in a territory — otherwise we would be bidding you against each other on storm night, which is the exact problem we sell against
- You want leads bought and resold to you. We never touch your leads
What will this cost me, honestly?
Below the gross profit on one storm roof.
A replacement in North Carolina averages just under $15,000 at 30–40% residential gross, so the monthly sits well under a single job. You pay Google directly and we never touch the spend — it is episodic by design, so a quiet month spends near nothing.
Exact numbers on the call — they depend on how many counties you cover, and quoting a figure before we have looked at your territory would be a guess.
What the agreement says
- Managed service, not software you have to run. We build and pre-approve the campaigns so they are ready to switch on.
- One contractor per trade in your territory — the ZIP codes you actually run trucks to. In writing.
- The rate you sign at is the rate you keep. It rises for the next contractor as territories fill, never for you.
- No lead fees, no commission, no share of the job. A flat monthly and your own ad spend.
Two ways to start.
Both free. Take whichever suits the morning you're having.
The ZIP check is instant and shows you three years of your county's storm record. The call is twenty minutes, by phone. One contractor per trade, in the ZIP codes you run.