The question we asked
“Who is the best landscaper in Prescott Valley, AZ? List 3-5 specific businesses by name with a short description of each.”
That is the exact prompt we sent to ChatGPT. One question, no follow-ups, phrased the way a homeowner with a torn-up yard would actually ask it.
This is worth testing because it is how a growing share of local search now works. Bain reported in 2025 that about 80% of consumers rely on AI-generated summaries for at least 40% of their searches, and roughly 60% of searches now end without a click to any website. The old pattern — scroll the results, compare a few sites, read some reviews — is being replaced by a shorter one: ask a question, get a shortlist, call someone on it. If your business is not on the shortlist, you were never in the running, and you will never know the customer existed.
What ChatGPT said
ChatGPT opened with a standard disclaimer that it cannot give real-time endorsements. Then it did what we asked and named five businesses, each with a specialty attached.
Watters Garden Center Landscaping Services was described as a northern Arizona name with custom, water-wise landscape design suited to the local climate. Native Creations Landscaping was credited for Arizona-native plants and eco-friendly work. Goodman’s Landscape Maintenance Services was positioned around ongoing maintenance, irrigation installation, and xeriscaping. Prescott Dirt Landscaping was called out for hardscaping — patios, walkways, retaining walls — plus full design and construction. Pinnacle Landscaping and Maintenance got a general design-build-and-maintain profile with a nod to craftsmanship and customer satisfaction.
The response closed by suggesting the reader contact these companies directly. Which means the shortlist was already set. Everyone else in Prescott Valley was competing for a call from a list they were not on.
Why these businesses got recommended
ChatGPT is not driving around Prescott Valley judging paver joints. It recommends businesses it has repeatedly seen described in the text it learned from and the sources it retrieves. Princeton’s GEO study (KDD 2024) showed that generative engines systematically favor content that is specific, quotable, and backed by recognizable sources — and that businesses represented that way get pulled into answers.
The signals that likely did the work here:
- Directory and profile density. Consistent listings with matching name, service list, and service area across the major local directories give the model multiple confirming sources.
- Review volume. A deep review history is third-party text describing what a business does well, in customers’ own words — exactly the material AI answers are built from.
- Brand mentions beyond their own websites. Local press, community forums, supplier and industry pages. A business that only describes itself is a single source; a business others describe is corroborated.
- Content depth. Service pages that actually explain water-wise design or retaining-wall construction give the model concrete language to reuse. Notice that ChatGPT’s descriptions read like compressed service pages — that is where they come from.
What the recommended businesses have in common
Four patterns stand out across the five winners:
- A legible specialty. Water-wise gardens. Native plants. Maintenance and irrigation. Hardscape. Each business compresses to one clear sentence. “Full-service landscaping for all your needs” gives a model nothing to grab.
- Climate-specific vocabulary. Xeriscaping, water-wise, Arizona-native. These are the terms people in central Arizona actually use when they write about landscaping, so businesses using them appear in the conversations the model learned from.
- A presence other people wrote. All five appear in text they did not author — reviews, directories, local mentions.
- Time in market. These are established names with years of accumulated citations. AI visibility compounds the same way reputation does.
What’s missing from the landscapers who weren’t recommended
There are far more than five landscapers serving Prescott Valley. The ones left out generally share the same gaps: a thin website that lists services without describing them, few or stale reviews, no mentions anywhere they did not write themselves, and generic copy that could belong to a landscaper in any state.
Princeton’s findings say these gaps are fixable — and quantify the payoff. Adding relevant statistics to content improved visibility in generative-engine answers by roughly 41%. Citing credible sources improved it even more — up to 115% for sites that were not already ranking well. Read that second number again: the biggest gains went to the businesses that were losing. This is not a game where the leaders are locked in.
What this means for your business
If you run a landscaping company in the Prescott Valley area, the checklist writes itself:
- Pick the specialty you want to be known for and state it identically everywhere — website, profiles, directories.
- Rewrite your service pages to actually describe the work, using the region’s vocabulary: xeriscaping, drought-tolerant planting, monsoon drainage, decomposed granite.
- Put verifiable specifics in your content — years in business, project scope, real numbers you can stand behind.
- Grow reviews continuously, and get mentioned in sources you do not control: local publications, supplier pages, community sites.
None of this is exotic. It is the difference between being described once, by yourself, and being described many times, by others.
Want to see your score?
Want to know where you stand? Book a 30-minute call. We ask the AI engines about your business while you watch, and you leave with next steps.