Most solar marketing advice is written for residential. Get on the comparison sites, chase reviews, win the quote race. For commercial solar, almost none of that is how buyers actually find you.
In August 2026 we ran six commercial solar queries through Google, Google AI Mode and ChatGPT, alongside ten residential ones for comparison. Things like “commercial solar installer melbourne” and “commercial solar for warehouses australia”.
In commercial solar, ChatGPT built 97% of its citations from installers’ own websites. Comparison sites were 3%.
In residential, aggregators were 35%. The two markets are barely the same channel.
Key findings
- Aggregators are almost absent from commercial answers. 3% of ChatGPT citations, against 35% residential.
- AI reads more of your site, not less. ChatGPT cited 5.5 sources per commercial answer against 2.6 residential.
- The field is small. Only 23 distinct installer domains were cited across all six commercial queries, against 43 across ten residential ones.
- Google organic is also installer-heavy in commercial: 86% installer sites, 14% aggregators.
- No single commercial installer dominated. The most cited appeared three times.
Residential and commercial are different games
| Residential | Commercial | |
|---|---|---|
| Aggregator share, ChatGPT | 35% | 3% |
| Installer sites, ChatGPT | 65% | 97% |
| Aggregator share, Google organic | 30% | 14% |
| ChatGPT sources per answer | 2.6 | 5.5 |
| Google AI Mode sources per answer | 6.4 | 4.7 |
| Distinct installers named | 43 | 23 |
Two numbers matter most here.
97%. In commercial, your website is not one input among many. It is very nearly the whole input. There is no SolarQuotes standing between you and the answer, which we covered in this study of residential solar.
23. Only 23 companies were named at all. That is a small field for a whole country. In residential you are fighting hundreds of installers and a wall of comparison sites. In commercial you are competing with a couple of dozen firms that have a website good enough to be quoted.
Why the field is so thin
Commercial buyers ask specific questions. Warehouse roofs, three-phase supply, payback periods, tender processes, kilowatt ranges. An AI answering those needs a page that talks about them concretely.
Most solar websites are built for homeowners. They talk about bills, rebates and roof types. When the question is a 200kW warehouse install, those pages have nothing to offer, so the model skips them and takes whoever does have a page on it.
That is the whole opportunity. The barrier is not authority or backlinks. It is that almost nobody has written the commercial pages properly.
What actually gets cited
Looking at the commercial installer sites that were cited, the pattern was consistent. The pages that got quoted carried specifics:
- System sizes in kilowatts, with real ranges rather than “systems of all sizes”
- Named case studies, ideally with the site type and the size installed
- Payback periods and an indication of cost, even as a range
- Which sectors they work in, named: warehousing, manufacturing, aged care, agriculture
- Whether they handle grid applications, tenders and after-install monitoring
- Accreditation in plain text, which ChatGPT checked in every solar answer we ran
What did not get cited was the homepage of a residential installer with a “commercial” item in the nav and a single thin page behind it.
What to do
- Build a real commercial section, not a page. One page per sector you actually serve, each naming the sector, typical system sizes and a case study.
- Put numbers on it. kW ranges, payback periods, a price range. Only 39% of Australian business homepages carry any price signal, and in a market of 23 competitors that is a cheap edge.
- Publish case studies with detail. Site type, size, outcome. These are what a model quotes when it explains why it named you.
- State your accreditation in text. SAA accreditation and NETCC approval, written out, not as a badge image. See what AI checks first.
- Do not bother chasing the aggregators here. They are 3% of the answer. Spend that effort on your own pages, where 97% of it lands.
- Check an AI crawler can fetch your site. Roughly one Australian site in eight returns an error to AI crawlers while serving browsers normally.
Residential solar marketing is about being included in someone else’s list. Commercial solar marketing is about having the only page that answers the question. The second one is more work and considerably more defensible.
If you want your commercial pages measured against the 23 firms currently being named, that is what our answer engine optimisation work starts with. Related: SEO for solar companies.
Common questions
How do commercial solar buyers find installers in Australia?
Increasingly by asking an AI a specific question, and the AI answers from installers’ own websites. In our test 97% of ChatGPT’s commercial citations were installer sites, against just 3% comparison sites.
Are solar comparison sites worth it for commercial work?
On this data, not really. They were 3% of ChatGPT’s commercial citations and 14% of Google’s organic results. For residential they matter far more.
How competitive is commercial solar in AI search?
Less than you would expect. Only 23 Australian installer domains were cited across all six commercial queries we ran. The field is thin because few firms have proper commercial pages.
What should a commercial solar page include?
System sizes in kW, sectors served by name, case studies with site type and size, an indication of payback and cost, and your accreditation written in plain text.
Does my residential reputation help my commercial visibility?
Not much directly. The AI answers commercial questions from pages that address commercial specifics. Reviews and comparison-site presence carry far less weight here.
Is commercial solar worth targeting if I mostly do residential?
If you can genuinely deliver the work, the visibility barrier is low right now. Job values are higher and the number of firms being named is small. If you cannot deliver it, publishing pages about it will only waste enquiries.
Method: six commercial and ten residential Australian solar queries run on 26 August 2026 through Google organic (desktop, top 20), the Google AI Mode SERP API and ChatGPT with web search enabled, location Australia. Citations counted at domain level and deduplicated per answer. This is a small sample from one day and AI answers vary between identical runs, so treat the proportions as indicative rather than precise.
