An HVAC contractor's website gets one job done well or it doesn't. Someone's furnace quit in January, they pick up a phone, search "furnace repair near me," and tap whichever company looks real and answers first. If the site is slow, vague about what you service, or buries the phone number, that person is already calling the next company on the list.
I'm in the middle of an HVAC contractor website rebuild right now, and I'm going to be careful in how I describe it. I won't name the company, and I'm not going to claim results, because the project isn't finished. What I can do is explain the decision, because it's the same one a lot of trades owners are facing: patch the site you have, or start over.
Why I didn't patch the old site
Patching sounds cheaper. Swap the photos, rewrite the home page, add a few service pages. Sometimes that is the right call, and I'll tell an owner so when it is.
It stopped being the right call here because of what the site needed to do next. The goal was never a prettier brochure. The goal was a site that AI features could be built on top of, features meant to bring in more customers. And that is a different requirement than "looks current."
Here's the problem with patching for that goal. An AI feature needs to read things and write things. It needs to know which services the company actually offers, what areas they cover, and when someone can be scheduled. Then it needs somewhere to send what it learns, like a new request that lands in the company's own records instead of dying in a form notification. If the old site stores everything as loose page content on a platform you can't get underneath, you end up gluing features onto a foundation that was never built to carry them. I've written about why that ceiling is structural on page builders, and the HVAC case isn't special. It's the same ceiling.
So the rebuild is the groundwork. I'm doing the part that makes later work cheap and possible, before building any of the later work.
What "ready for AI" meant in practice
I wrote a whole post on what makes a website AI-ready, so I'll keep this to what it looked like on a trades site specifically.
Each service gets its own page. Furnace repair, AC repair, installs, maintenance plans, and whatever else the company really does each get a page that says plainly what's included and where. A single "Services" page with six paragraphs on it is hard for a customer to skim and hard for software to read accurately.
The basics exist as real text and as structured data in the code. Name, phone, service area, hours, emergency availability. If any of that is trapped inside an image or a PDF, a search engine or an AI assistant can't use it.
The site is built so other tools can connect to it. The code and the data belong to the owner, which means a missed-call text, a quote request reply, or a scheduling step can be attached later without asking a platform for permission. That last point is the reason I build custom, and it's also why owning your website matters more than it sounds like it should.
None of this is visible to a customer on day one. They just see a fast site that answers their question. The value shows up when the next piece gets added.
What the owner has to provide
This is the part that scares people, so I'll be specific about how I run it with any owner, not only this one. I draft the pages. The owner corrects them. I ask for the things only they know: what jobs they want more of, what they won't touch, which towns they really serve versus which ones they'd like to, and real photos of their trucks, techs, and finished work. Stock photos of a smiling man holding a wrench don't build trust with anyone.
I also need access to the tools they already use, since the AI part only helps if it connects to something real. That conversation happens early, not at the end. The full walkthrough of how a project runs is in what actually happens when I rebuild a small business website, and it covers the timeline and what I need from you week by week.
What I'm not saying
I'm not saying every HVAC company needs a rebuild. If your site loads fast, your pages are clear, and you have no plans to add anything to it, leave it alone. A rebuild has a real cost, and I've laid out what custom sites actually cost so you can judge whether the math works for your shop.
I'm also not saying the AI features are done. They're being implemented, and I'll write about what works and what doesn't once I have something honest to say. For now the accurate version is this: I'm laying a foundation so that those features can be built, and I'd rather tell you that than overstate where things are.
If you run a heating and cooling company and you're wondering whether your current site can carry anything beyond a phone number and a contact form, that's a fair question to ask before spending money. Asking it is free.
If you want to talk through what this would look like for your business, the audit is free and takes 30 minutes. Get in touch