The site comes first. We replace it before any other work starts.

Home-service companies get a four-week rebuild on a template written for answer engines. Schema, direct answers, a single business identity, and a structure assistants can cite are part of the build, not a later add-on.

Every package includes this rebuild. It is not optional and it is not a line item we drop.

It is a full reconstruction of the company website on a base that answer engines can parse, quote, and attribute to one business.

The build covers schema, pages that open with the answer, a consistent definition of the company, headings written as the questions people ask, and the technical setup AI crawlers need.

A redesign changes how the site looks. This changes how a machine understands it. Schema, structured data, and answer-shaped copy do not sit reliably on top of a theme that was built as a brochure. That is why the site is replaced before optimization begins.

The site a home-service company has today was almost always made for a person reading it. It can look right, load quickly enough, and describe the work. For a customer browsing, that can be plenty. For AI search, it is not.

ChatGPT, Google AI Mode, Google AI Overviews, and Gemini do not scan a page the way a homeowner does. They pull out entities, read schema, look for a sentence that answers the question, and check that story against other sources. A site written only for people gives them very little to lift. A site written for extraction gives them a citation they can stand behind.

Patching came first in our own tests. Schema was added to old themes. Copy was rewritten in place. Entity notes were tucked into metadata. Rebuilding produced better results than those patches, and the gap was consistent.

The reason is structural. An existing site carries limits we do not control: themes that strip schema, a CMS that warps structured data, templates that resist a direct answer, and an old URL history that makes one company look like several. Optimization on that base flattens out.

The replacement is a template made for home-service answer-engine work. The schema is right. The pages follow the answer pattern those systems cite. The entity record does not contradict itself. The base is built once. After that, monthly work adds to it instead of repairing it.

The scope does not shrink to hit a lower price, and nothing on this list is sold as an extra. Each piece is required for the result. Every client receives the full set.

$5,000

$2,500 when you sign, with the first month. $2,500 when we ship the site.

That fee covers up to 30 templated pages: home, about, contact, service pages, and service-area pages. The page count is taken from your sitemap before you sign. It is not discovered halfway through the build. Blog articles belong to the monthly content plan and do not count against the 30. Any page past the first 30 is $100.

Every package adds the same kind of ongoing work: answer-engine articles each month, Google Business Profile management, review responses the day they come in, directory and authority citations, weekly prompt tracking across ChatGPT, Google AI Mode, Google AI Overviews, and Gemini, and a monthly AI visibility diagnostic.

The rebuild runs four weeks from the signed contract to launch. The first two weeks are discovery, strategy, and the writing. Week three is the build, schema, and quality checks. Week four is launch prep, indexing, and the handoff into the six-month program.

Packages are a six-month term, billed monthly, plus the $5,000 build. If the current site was built before 2024, four weeks of rebuilding has moved measurable visibility further than continued optimization on the old site. That conclusion comes from a year of comparing the two approaches.