Your website is either a salesperson or a brochure.

One of them books estimates while you sleep. Here's the structure that separates them, and how to tell which one you own.

The short version

A contractor website generates leads when it is built as a booking machine rather than a portfolio: a dedicated page for every service and every city you serve, proof visible before the pitch, one call to action repeated everywhere, fast pages on mobile, schema markup underneath, and follow-up that fires within a minute of the form landing.

Most contractor sites fail on the first item alone. One page trying to rank for twelve services across twenty towns ranks for none of them.

What actually separates the two.

Six things. None of them are about how the site looks, which is the part most contractors are sold on.

A page per service, a page per city

Google ranks pages, not companies. AI assistants read them the same way. A homeowner searching one specific job in one specific town should land on a page about exactly that.

This is the whole ballgame and it is where nearly every contractor site is thin. A services page listing twelve offerings competes for none of them. Twelve pages, each written about one job, each naming the trade and the territory, compete for twelve.

The same applies to geography. A footer stuffed with town names is not city coverage; it is a signal that somebody knew city pages mattered and did the cheap version. Real coverage means a page per city with content that could only have been written about that city.

Done properly this adds up fast. Rock N Roll Stoneworks' site runs 70+ pages, one for every service, product line, and Front Range city they cover, and it pulls 20+ leads a month on its own with no ad spend behind those visits.

Proof before the pitch

Homeowners are hiring strangers to tear into their house. The site's first job is making you not a stranger.

Reviews with real names, license numbers, manufacturer certifications, and photography from your own jobsites. Not stock images of kitchens you did not build, which experienced homeowners spot immediately and which quietly costs you more than having no photos at all.

Position matters as much as presence. Proof placed below the fold, on a separate testimonials page nobody visits, is proof you paid for and did not use. It belongs next to the request-an-estimate button, where the hesitation actually happens.

Specific beats impressive. A named homeowner describing the job you did in the town they live in outperforms a five-star average with no detail attached.

One action, everywhere

Every page drives to the same next step: request the estimate. Five competing buttons convert worse than one clear one.

Call now, text us, email us, download the guide, follow us, request a quote. Every additional option is a decision the homeowner has to make before doing the thing you actually want, and each one costs conversions.

On mobile the phone number should be one tap, always visible, never a graphic. A meaningful share of contractor traffic is a homeowner standing in the room they want changed.

The form itself should ask for the minimum that lets you qualify. Every extra field is another chance to abandon, and the details you skipped are what the qualifying call is for.

Speed and mobile, which are the same problem

Most homeowners meet your company on a phone screen at night. A slow site reads as a sloppy crew, fair or not.

Page speed is a ranking factor, but the conversion cost lands first. A homeowner comparing three contractors on a couch does not wait for a hero video to decode.

The usual culprits are uncompressed project photography, a page builder loading scripts nothing uses, and embedded widgets nobody chose deliberately. Contractor sites are image-heavy by nature, which makes discipline here worth more than it is in most categories.

Test on a phone on cellular data, not on the office wifi on a desktop. That is the condition your buyer is actually in.

The invisible layer

Schema markup states in machine-readable form who you are, what you do, where you work, and what your reviews say. Prose can be read several ways. Schema cannot.

This is the layer that turns a website into an asset AI assistants can use. When a homeowner asks an assistant who to hire, the answer is assembled from sources it can verify, and your site is the primary one.

It is also among the cheapest work on this list. Schema is code, not content, and it goes in once rather than being maintained forever.

It is the bridge between this page and answer engine optimization, which is the same structural work aimed at a different destination.

What happens after the form

A form landing in an inbox nobody checks is a beautiful way to pay for your competitor's appointment.

The site should feed a CRM that fires the first touch within a minute. Past a minute, conversion falls off a cliff, and the homeowner who filled your form filled two others in the same sitting.

This is the part that gets skipped when a website is bought as a standalone project. The site is scoped, built, launched, and handed over, and the follow-up is assumed to be somebody's job. It is nobody's job by default.

Judge a website proposal by what it says about minute one after the form, not by the design mockups.

Audit your own site in ten minutes.

No vendor required. Run these and count how many you fail.

Count your service pages. One page per service, or one page listing them all?
Count your city pages. Real pages, or town names in the footer?
Open your homepage on a phone on cellular data and time it.
Find the request-an-estimate button. How many other buttons compete with it?
Look for proof above the fold: named reviews, license number, your own photography.
Submit your own form and time how long until something reaches you.
Search your business name and see whether the description Google shows is one you wrote.

20+/mo

Leads from Rock N Roll Stoneworks' website alone, before a dollar of ad spend, across a 70+ page build.

Read the case study

“Get this: a homeowner asked AI who the best company to build his backyard pool was and it said Rock N Roll Stoneworks. That's wild. He signed today.”
Jordan Jackson · Owner, Rock N Roll Stoneworks
The same structural work is why an AI assistant had something to recommend when a homeowner asked who should build his pool. A site built to be read by machines gets read by both kinds.

What a new website won't do.

A site is the foundation the other layers sit on. Foundations don't generate traffic by existing.

It won't fill the calendar by itself
A new site with no traffic is a better brochure. Rock N Roll's 20+ leads a month come from a site that also has SEO, AEO, and an active Google Business Profile behind it. Paid ads are what carry the calendar during the 2 to 4 week build.
It won't rank on day one
Structure is what makes ranking possible; it isn't ranking. The organic layers compound over 3 to 6 months depending on your market and trade, which is the honest timeline and the reason both layers run together.
It won't fix slow follow-up
If nobody answers inside a minute, a better site just means paying more per lost appointment. The CRM and the speed-to-lead handoff are part of the build for exactly this reason.

Straight answers on contractor websites.

Why isn't my contractor website generating leads?: Usually structure. One generic page trying to rank for every service in every city ranks for none of them. The fix is dedicated service and city pages, proof visible before the pitch, one clear call to action, and follow-up within a minute of the form landing.
How many pages should a contractor website have?: One per service and one per city you serve, plus proof and process pages. Rock N Roll Stoneworks runs 70+ pages and gets 20+ leads a month from the site alone. Depth is what ranks and what AI assistants read.
How long does a contractor website take to build?: 2 to 4 weeks depending on scope, when it's built by a team that ships weekly. Paid ads carry the calendar while it's in build, and SEO and AEO compound on top of it over 3 to 6 months.
What does a contractor website cost?: Here it isn't sold as a standalone project; the site is part of the engagement and scoped with the rest of it around your market and trade. That's deliberate: a site handed over without the traffic and follow-up behind it is the most common way contractors waste money on marketing.
Should I use Wix, Squarespace, or a page builder?: They can produce a presentable site. Where they tend to struggle is depth and speed: dozens of service and city pages with clean structure and schema, loading fast on cellular. If your plan is a five-page site, a builder is fine. If your plan is to rank, the architecture matters more than the tool.
Do websites still matter now that AI answers questions directly?: More than before. Assistants assemble answers from sources they can read and verify, and your website is the primary one. A structured site with schema and real proof is exactly what gets you named in an AI answer rather than skipped.
How do I know if my site is fast enough?: Open it on a phone on cellular data, not on office wifi on a desktop, and count. That's the condition your buyer is in. If you want a number, run it through Google's PageSpeed Insights, but the phone test tells you most of it.
Do I need city pages if I only serve one metro?: Yes, usually more than a multi-metro contractor does. Homeowners search by suburb, not by metro. A page per suburb you actually work in is often the difference between showing up for a search and not existing for it.
Should the website be part of the marketing engagement or bought separately?: Separately is where most of the waste happens: the site gets scoped, built, handed over, and then nobody owns whether it produces. We build it as part of the engine because the traffic, the follow-up, and the site have to be one decision to work.

Want a site built to book, not to admire?

It ships in 2 to 4 weeks, with paid appointments carrying the calendar while it's in build. One contractor per trade, per market, so the first thing we check is whether yours is open.

Check if your market's open See the sites we've built

What actually separates the two.

Six things. None of them are about how the site looks, which is the part most contractors are sold on.

A page per service, a page per city

Google ranks pages, not companies. AI assistants read them the same way. A homeowner searching one specific job in one specific town should land on a page about exactly that.

Proof before the pitch

Homeowners are hiring strangers to tear into their house. The site's first job is making you not a stranger.

Specific beats impressive. A named homeowner describing the job you did in the town they live in outperforms a five-star average with no detail attached.

One action, everywhere

Every page drives to the same next step: request the estimate. Five competing buttons convert worse than one clear one.

On mobile the phone number should be one tap, always visible, never a graphic. A meaningful share of contractor traffic is a homeowner standing in the room they want changed.

The form itself should ask for the minimum that lets you qualify. Every extra field is another chance to abandon, and the details you skipped are what the qualifying call is for.

Speed and mobile, which are the same problem

Most homeowners meet your company on a phone screen at night. A slow site reads as a sloppy crew, fair or not.

Page speed is a ranking factor, but the conversion cost lands first. A homeowner comparing three contractors on a couch does not wait for a hero video to decode.

Test on a phone on cellular data, not on the office wifi on a desktop. That is the condition your buyer is actually in.

The invisible layer

Schema markup states in machine-readable form who you are, what you do, where you work, and what your reviews say. Prose can be read several ways. Schema cannot.

It is also among the cheapest work on this list. Schema is code, not content, and it goes in once rather than being maintained forever.

It is the bridge between this page and answer engine optimization, which is the same structural work aimed at a different destination.

What happens after the form

A form landing in an inbox nobody checks is a beautiful way to pay for your competitor's appointment.

The site should feed a CRM that fires the first touch within a minute. Past a minute, conversion falls off a cliff, and the homeowner who filled your form filled two others in the same sitting.

Judge a website proposal by what it says about minute one after the form, not by the design mockups.

Audit your own site in ten minutes.

No vendor required. Run these and count how many you fail.

Count your service pages. One page per service, or one page listing them all?

Count your city pages. Real pages, or town names in the footer?

Open your homepage on a phone on cellular data and time it.

Find the request-an-estimate button. How many other buttons compete with it?

Look for proof above the fold: named reviews, license number, your own photography.

Submit your own form and time how long until something reaches you.

Search your business name and see whether the description Google shows is one you wrote.

What a new website won't do.

A site is the foundation the other layers sit on. Foundations don't generate traffic by existing.

It won't fill the calendar by itself

A new site with no traffic is a better brochure. Rock N Roll's 20+ leads a month come from a site that also has SEO, AEO, and an active Google Business Profile behind it. Paid ads are what carry the calendar during the 2 to 4 week build.

It won't rank on day one

Structure is what makes ranking possible; it isn't ranking. The organic layers compound over 3 to 6 months depending on your market and trade, which is the honest timeline and the reason both layers run together.

It won't fix slow follow-up

If nobody answers inside a minute, a better site just means paying more per lost appointment. The CRM and the speed-to-lead handoff are part of the build for exactly this reason.

Straight answers on contractor websites.

Why isn't my contractor website generating leads?

Usually structure. One generic page trying to rank for every service in every city ranks for none of them. The fix is dedicated service and city pages, proof visible before the pitch, one clear call to action, and follow-up within a minute of the form landing.

How many pages should a contractor website have?

One per service and one per city you serve, plus proof and process pages. Rock N Roll Stoneworks runs 70+ pages and gets 20+ leads a month from the site alone. Depth is what ranks and what AI assistants read.

How long does a contractor website take to build?

2 to 4 weeks depending on scope, when it's built by a team that ships weekly. Paid ads carry the calendar while it's in build, and SEO and AEO compound on top of it over 3 to 6 months.

What does a contractor website cost?

Here it isn't sold as a standalone project; the site is part of the engagement and scoped with the rest of it around your market and trade. That's deliberate: a site handed over without the traffic and follow-up behind it is the most common way contractors waste money on marketing.

Should I use Wix, Squarespace, or a page builder?

They can produce a presentable site. Where they tend to struggle is depth and speed: dozens of service and city pages with clean structure and schema, loading fast on cellular. If your plan is a five-page site, a builder is fine. If your plan is to rank, the architecture matters more than the tool.

Do websites still matter now that AI answers questions directly?

More than before. Assistants assemble answers from sources they can read and verify, and your website is the primary one. A structured site with schema and real proof is exactly what gets you named in an AI answer rather than skipped.

How do I know if my site is fast enough?

Open it on a phone on cellular data, not on office wifi on a desktop, and count. That's the condition your buyer is in. If you want a number, run it through Google's PageSpeed Insights, but the phone test tells you most of it.

Do I need city pages if I only serve one metro?

Yes, usually more than a multi-metro contractor does. Homeowners search by suburb, not by metro. A page per suburb you actually work in is often the difference between showing up for a search and not existing for it.

Should the website be part of the marketing engagement or bought separately?

Separately is where most of the waste happens: the site gets scoped, built, handed over, and then nobody owns whether it produces. We build it as part of the engine because the traffic, the follow-up, and the site have to be one decision to work.