The hail season is won by whoever answers the insurance question.
A storm-response contractor in the hail belt, competing against out-of-state chasers who arrive within days of every event.
Across one storm season.
Three figures from the engagement, each traced to the client's own reporting rather than ours.
Inbound claim enquiries
Form submissions that mention insurance or a claim, compared with the equivalent storm season the year before.
Time to first enquiry
After the season's first significant hail event, the first inbound enquiry came in 14 hours. The previous year the wait was eight days.
Claims content citations
The claims and insurance query set, where the site is now the source AI assistants quote for homeowners in the region.
FIGURES APPROVED IN WRITING BY THE CLIENT. COMPANY NAME WITHHELD UNDER A CONFIDENTIALITY AGREEMENT.
The situation
In the hail belt the work arrives in bursts, and out-of-state crews arrive with it. This contractor was local, licensed, and losing the first two weeks of every season to companies that knocked doors faster. The website had nothing at all about insurance claims, which is the only thing a homeowner researches after a storm.
What the audit found
Twenty-four distinct claim-related questions had real search volume in the region and no adequate local answer published by anyone. Deductibles, supplements, adjuster disputes, ACV against replacement cost. The national brands answered generically and no local contractor answered at all.
What we did about it
A claims section built as an answer set. One page per question, each with the direct answer first, state-specific rules named, and no gate in front of it. Storm-event pages that can be published within hours of an event, with the date, the affected suburbs, and what to do in the first 48 hours.
Where it landed
The claims pages did the work before the season started. By the time the first event hit, the site was already the thing being quoted when homeowners in the region asked what their policy covered.
Before the season, and during it.
The content was published four months before the first storm. That is why it worked.
Nothing published about claims
The regional claims answer set
FIGURES SUPPLIED BY THE CLIENT AND APPROVED IN WRITING. COMPANY NAME WITHHELD UNDER A CONFIDENTIALITY AGREEMENT.
Built in the off-season, on purpose.
Paid Discovery
A 24-question claims query set, regional competitor coverage analysis, and a storm-event page template in the blueprint.
Claims answer set
One page per question with state-specific rules, plus service-area pages for the suburbs the hail history says get hit.
Ready before the season
Published four months ahead of the first event, so the pages had time to be indexed and cited before they were needed.
Read another one.
Three more engagements, each with the figures approved in writing by the client.
Invisible in his own city
Twenty-two years in business, cited in no AI answer about roofing in a metro of 420,000. Twelve months after launch the site is the quoted source and the ad budget is half what it was.
Ninety-six pages, then thirty-four
Four metros under one brand, competing against themselves with near-identical service-area pages. Fewer pages, written per market, ended the cannibalisation.
Six-figure work sold from a homeowner site
Facility managers research materials, systems and lifecycle cost long before contractors. A specification answer set took commercial enquiries from four a quarter to seventeen.
Questions about this engagement.
Is claims content risky to publish?+
It is reviewed for accuracy against state rules and it never gives legal advice. It explains process, which is what homeowners are actually searching for.
How fast can a storm page go live?+
Same day. The template is built during the rebuild, so publishing is filling in the event, the suburbs, and the date.
Does this work outside the hail belt?+
The mechanism does. The questions change with the weather and the state rules.