How to get a solar company recommended by ChatGPT, Perplexity and Google AI answers
What AI answer engines use to name local solar installers, the page structure they quote, the review language that feeds them, and a 30-day setup plan.
A homeowner in your county types "who is a good solar installer near me that does not use door to door sales" into an assistant. The answer names three companies, gives one sentence about each, and the homeowner calls one of them. No search results page, no map pack, no ad auction.
That is already happening for solar more than for most trades, because solar is exactly the kind of expensive, confusing, high-anxiety purchase people ask an assistant to help them think through. Questions like which brands are worth paying for, whether a lease is a bad idea, and what a fair price per watt is all get asked in chat now, and the assistant frequently names installers while answering.
Getting named is not a trick. It is a specific set of things that are individually boring.
What the answer engines actually pull from
Four sources, roughly in order of weight for a local company.
Your website text, if it is quotable. These systems lift sentences. A page that says "We install residential solar and battery storage in Maricopa and Pinal counties and have done so since 2014" is a liftable sentence. A hero that says "Powering a brighter tomorrow" is not, and a page whose facts are only inside a slider or a PDF may as well not exist.
Your Google Business Profile. Category, service area, hours, review count and rating flow into several assistants either directly or through the search index they call.
Reviews, especially specific ones. An assistant summarizing local installers weights what reviews say, not just how many. Reviews that name the system size, the town and the outcome give the model something concrete to repeat.
Third-party mentions. Local news, a chamber listing, a county contractor list, a supplier's dealer locator, an industry association directory. These are how a model confirms you are a real entity rather than a website.
Fix the entity facts first
Every assistant needs to be sure you are one company. Solar businesses are unusually bad at this because they rebrand, get acquired, run DBAs, and end up with three variations of their name in circulation.
Pick one exact legal or trading name, one address format, one phone number, and one URL. Then make them identical on your website footer, your Google profile, your Facebook page, your state contractor license record, your supplier dealer pages, and every directory that lists you. Set aside an afternoon and fix all of them at once. Inconsistency here is the most common reason a real company gets skipped in an AI answer.
Add Organization schema and LocalBusiness schema to the site with the same name, address, phone, founding year and a sameAs list linking your profiles. This is not a ranking hack. It is a machine-readable statement of the same facts, and it removes ambiguity.
Write pages a machine can quote
Structure every important page the same way:
- An H1 that names the service and the place.
- A first paragraph, two or three sentences, that states plainly what you do, where, since when, and one differentiating fact.
- H2 sections written as questions a homeowner would ask, answered immediately underneath in 40 to 80 words.
- A short specifics block: brands installed, warranty terms, typical timeline from signature to permission to operate, financing options, counties served.
- Real photos with descriptive file names and alt text.
The pages worth building this way for a solar company: residential solar installation, battery storage, solar panel repair and service, solar removal and reinstall for roofing, commercial solar if you do it, one page per county or major city you install in, a financing page, and a page that explains your warranty and service process in detail.
That last one is unusual and it is the one assistants quote most often, because "what happens if something breaks in year ten" is the question homeowners ask and almost no installer answers in writing.
Answer the awkward questions in public
Assistants are asked skeptical questions about solar constantly. Whether it is worth it, whether leases are a scam, whether the panels damage the roof, whether the company will still exist. Whoever answers those honestly in indexable text is who gets quoted.
Write short, direct pages for the ones you can answer with integrity:
- Is solar worth it on a house with a north-facing roof
- Lease, PPA, loan and cash compared, with the drawbacks of each stated
- What happens to your solar warranty if the installer goes out of business
- How roof penetrations are flashed and what your leak warranty covers
- What to ask any solar company before signing
Do not write these as sales pages. Write them the way you would explain it at a kitchen table when you already know the homeowner is not buying today. Include the cases where the answer is no. A page that says "if your roof has under seven years left, do not put solar on it, replace the roof first" gets quoted, because it is the kind of statement an assistant recognizes as trustworthy.
Reviews are training data now
This is where the reviews guide and this one meet. Ask installed customers to name the system size, the town and what their bill did. Ten reviews written that way do more for AI visibility than a hundred that say "great job."
Volume still matters as a credibility floor. Across 32,440 solar listings the median company has 18 reviews and 55.2% have fewer than 25. An assistant asked for the best installer in a county has to choose from something, and a company with 2 reviews rarely makes the list.
Do not block the crawlers
Check your robots.txt. Some site builders and security plugins block AI user agents by default, and some installers block them deliberately without realizing it removes them from AI answers entirely. If you want to be recommended, let the assistants read the site.
Add an llms.txt file at the root listing your main pages with one-line descriptions. It is a young convention, it costs an hour, and it helps.
Get mentioned somewhere that is not your own site
Five realistic targets for a solar company:
- Your equipment manufacturer's certified installer locator.
- Your state or regional solar association member directory.
- The local chamber of commerce.
- A local news story. Reporters cover utility rate changes and interconnection backlogs and need someone technical to quote. Email the business desk with a two-paragraph explanation of a real local change and offer to explain it.
- A supplier or financing partner case study.
Five solid citations beat fifty directory submissions.
A 30-day plan
- Week 1: standardize name, address, phone and URL everywhere. Add schema. Check robots.txt.
- Week 2: rewrite the home page and the two main service pages to the structure above. Publish the warranty and service page.
- Week 3: publish four of the awkward-question pages. Publish llms.txt.
- Week 4: change your review ask to request system size and town. Send the local reporter email. Claim the manufacturer locator listing.
Then test. Once a month, ask ChatGPT, Perplexity and Google's AI answers for solar installers in your city and write down who gets named. Watch it move over a quarter.
Frequently asked
Does this hurt my Google rankings? No. Every item on this list is standard technical and content quality work. The pages that get quoted also rank.
Is there a way to pay for placement? Not in the answers themselves. Anyone selling you AI ranking placement is selling something else.
How long until it changes anything? Six to twelve weeks for the entity and schema work, longer for the citations. It compounds and it does not reset when a platform changes its layout.
AI search optimization is one of the services in the free 14-day trial. Text or call (385) 832-6175. Related reading: the website essentials guide and the benchmark report.