How to Get Your Business Cited by ChatGPT and Google AI Overviews
Someone in your market asked ChatGPT for the best business like yours this week. It named a few. Were you one of them? This is the practical version: the exact moves that get a business pulled into answers from ChatGPT, Google AI Overviews, Perplexity, and Gemini. If you want the how-it-works theory first, we covered what AI SEO actually is separately. This post is the doing, and it starts with proof: three of our own clients that Google's AI is recommending by name right now.
In AI search there's no page two. You're either named in the answer or you don't exist for that question.
We've Done This. Here's the Proof.
Before any of the how-to, here's why it's worth reading. These are three of our clients, pulled up in a fresh incognito window. No login, no personalization, no cached history nudging the result. Just what a stranger sees when they ask Google. In every one, Google's AI Overview recommends our client by name, and the words it uses are lifted almost straight off the page we built.
Newman Electric, named for EV charger installation
Incognito search: "best ev charger electrician in vancouver wa." Google's AI Overview calls out Newman Electric as a top local option and describes them as licensed, bonded, permit-pulling contractors who install a range of charger types.
Incognito Google search captured June 29, 2026. No login, no personalization. Newman Electric is a Savo Group client.
What Google's AI said
"These licensed and bonded contractors service the entire Clark County area, handle necessary local permits, and install a variety of vehicle charger types."
What's on the page we built
"Newman Electric has installed 120+ Level 2 EV chargers across Vancouver, WA and Clark County since 2021. We run the dedicated 240V circuit, mount the unit, pull the L&I permit, and make sure everything passes inspection."
That's not luck. The page answers the exact question a buyer is asking, with real specifics, and it's backed by Electrician schema that spells out the service area as Clark County and even lists "Clark County electrical permits" as something the business knows. The AI didn't have to guess what Newman does or where. We told it, in plain text and in code.
Sniff and Go, named for private dog walking in San Francisco
Incognito search: "best private dog walking in san francisco, ca." The AI Overview lists Sniff and Go first and nails the thing that makes them different, down to the price.
Incognito Google search captured June 29, 2026. No login, no personalization. Sniff and Go is a Savo Group client.
What Google's AI said
"Specializes exclusively in private, on-leash walks in your dog's familiar neighborhood... trainer-led operations with W-2 employees rather than independent contractors."
What's on their page
"All private, all on-leash, all in the client's neighborhood." ... "W-2 employees with 40+ hours of paid training by CPDT-KA certified trainers, not a freelancer or gig worker."
It even quoted the prices: 28 dollars for a potty break up to 88 dollars for a 60-minute walk. Those numbers are on the page and inside an OfferCatalog in the schema, alongside an 11-question FAQ that lays out the W-2 versus contractor difference the AI repeated almost word for word.
AvilaCo Drywall, named first in a competitive market
Incognito search: "who's the best drywall installer in vancouver wa." Vancouver is a real market with real competition, and Google's AI Overview puts AvilaCo Drywall at the top of the list.
Incognito Google search captured June 29, 2026. No login, no personalization. AvilaCo Drywall is a Savo Group client.
What Google's AI said
"AvilaCo Drywall: A highly regarded, family-owned local contractor. They specialize in both residential and commercial projects, including new builds, renovations, popcorn ceiling removal, and decorative plaster."
What's in their schema
An OfferCatalog listing each service by name: Drywall Installation, Drywall Repair, Hand Texture, Decorative Plaster, Popcorn Ceiling Removal, and more.
Look at what the AI singled out: "popcorn ceiling removal" and "decorative plaster." Those are niche services most drywall sites bury or skip. They got quoted because each one is declared by name in AvilaCo's structured data. That's schema doing exactly what it's supposed to do.
From invisible to first-cited in one quarter
We started AvilaCo's marketing in March 2026. Before that, they were invisible. No first-page rankings for anything, basically no monthly organic traffic. Three to four months later, a stranger searching "who's the best drywall installer in vancouver, wa" gets AvilaCo as the very first name Google's AI hands them, and the number two business in the Map Pack for that same search. From nowhere to the top of the answer in a single quarter.
And it's not only the big market. Run the same kind of search for a small town nearby and AvilaCo shows up there too, because we built a dedicated page for it.
Incognito search for "best drywall installer in kalama wa," captured June 29, 2026. Same business, smaller town, same result.
Three businesses, three industries, four searches across two states, all from clean incognito windows. None of it was bought. No ad slot, no pay-to-play. The rest of this post is how it's done.
About these screenshots: All four were captured on June 29, 2026 in a Google incognito window with no account signed in, so the results aren't shaped by personal search history. If you run the same searches while logged in, Google personalizes what you see based on your location, history, and past activity, so your results may look different. AI Overviews and search rankings also shift over time. These are a snapshot of one day, not a fixed, permanent position.
Want to grow your business's AI visibility?
We'll show you exactly where you show up today and what it takes to get cited. Free, no pitch.
Start by Checking Where You Actually Stand
Before you change a single thing, go find out if you're already showing up. This takes twenty minutes and tells you more than most paid audits.
Open ChatGPT, Perplexity, and Google's AI Mode. Run the questions your actual customers would type. Not your branded name. The real ones:
- "Best [what you do] in [your city]"
- "Who should I call for [your service] near [your city]?"
- "How do I choose a [your type of business]?"
- "Is [your company name] any good?"
Write down what comes back. Which businesses get named. Which websites the AI cites as its sources. Whether your name shows up at all, and if it does, whether the AI got the details right. Put it in a simple sheet with a row per query so you have a baseline to measure against later.
The uncomfortable part
If you're not named and you don't recognize the sites the AI is citing, those sites are your competition for the answer. The goal of everything below is to become one of them.
Rank First, or Stay Invisible
Here's the thing most "AI SEO" pitches skip. AI models pull heavily from pages that already rank in regular search. If Google doesn't trust you, ChatGPT and AI Overviews mostly won't either. There is no shortcut that skips the foundation.
That doesn't mean you need to rank number one for the exact phrase. A December 2025 Surfer SEO study found that 68% of pages cited in AI Overviews weren't in the top 10 organic results for the original query. They ranked for related searches. The AI broke the question into pieces, searched each one, and pulled from a wide pool. So broad, healthy organic presence across your services and your area is what feeds the machine.
The foundation is boring and unavoidable. A fast, properly built site. A Google Business Profile that's actually filled out. Real reviews. Content covering each service you offer and each area you serve. If your local SEO is weak, fix that before you chase anything labeled "AI." Anyone selling AI SEO as a separate magic product, detached from the SEO you already need, is selling you a second invoice for the same work.
Why We Build Static HTML, Not WordPress
None of this works if the crawler can't read your page. And this is where a lot of businesses are quietly losing before they even start.
A static HTML site, the kind we hand-build, puts your words and your schema right there in the file the moment it loads. Nothing to execute, nothing to wait for. A crawler asks for the page and gets every sentence and every piece of structured data in one shot. The three sites above are all static. The content Google quoted was sitting in plain HTML, ready to grab.
Now picture the typical WordPress setup. A theme, a page builder, and forty plugins, all loading scripts, rendering your actual text with JavaScript after the fact. Google can sometimes run that JavaScript and figure it out, eventually. But it's slower, it's less reliable, and when a page is heavy enough, the crawler renders less of it or gives up. The AI crawlers behind ChatGPT and Perplexity are even less patient than Google. If your content only appears after a pile of scripts finishes running, there's a real chance the bot never sees it. Your page isn't just ranking lower. For that crawler, parts of it don't exist.
That's the whole reason we don't build bloated WordPress sites. A slow, plugin-stuffed site can be partly invisible to the exact systems you're trying to show up in, and most owners have no idea it's happening. We hand-code a fast static site instead, with the content and the schema baked straight into the HTML. Some of our clients came to us off a sluggish WordPress build, and the rebuild is often the single biggest jump in how much of their site search engines and AI tools can actually read. If you want the longer version of that argument, it's the backbone of how we approach web design.
What Makes a Page Quotable by an AI
A model citing you is really a model lifting a chunk of your page and dropping it into its answer. Your job is to make that easy. The original GEO research out of Princeton and Georgia Tech (KDD 2024) tested this directly and found that adding statistics, citations, and authoritative quotes boosted a source's visibility in AI answers by up to 40%. That tracks with what we see. Here's how to apply it.
- Answer the question in the first two sentences. Put the answer up top, then explain underneath. Models grab the clean, direct answer near a heading. If the reader has to scroll through three paragraphs of windup to find it, so does the AI, and it'll usually go grab a competitor instead.
- Use real numbers, dates, and named sources. "We offer fast, affordable service" is unquotable. "Most repairs run $150 to $400 and we answer same-day" is something an AI can actually hand to a user. Specific beats vague every time, and it's the single biggest lever the research found.
- Write in liftable chunks. One idea per section. The heading is the question, the paragraph right under it is the answer. Short, self-contained sections give the model something it can quote without dragging in noise.
- Show the work and date it. Back up claims with a source. Mark when the page was updated. Perplexity in particular favors fresh content, and an "Updated June 2026" line beats an undated wall of text.
- Cover the whole question, not one keyword. Because the AI fans your query out into a dozen sub-questions, one stuffed page won't cut it. You need pages across your services, your area, and the questions customers actually ask. We get into the why of that in the AI SEO explainer.
Give the Machine the Labels
Structured data, also called schema markup, is code that tells a machine exactly what your business is in a language it can't misread. It won't get you cited on its own. What it does is remove ambiguity, so the model is confident about what you do, where you are, and what each page is about.
For a local or service business, the schema types that earn their keep are Organization, LocalBusiness, Service, FAQPage, and BreadcrumbList. Mark up your business details, your services, and the common questions you answer. Then keep your name, address, and phone number identical everywhere they appear, on your site and off it. Inconsistent business info is one of the fastest ways to make a model distrust you.
Here's a trimmed piece of the actual structured data on AvilaCo Drywall's site, the client from the drywall example above. It's nothing fancy. It's a machine-readable statement of what they do and where:
{
"@type": "Service",
"name": "Drywall Contractor in Kalama, WA",
"serviceType": "Drywall Contractor",
"areaServed": [
{ "@type": "City", "name": "Kalama" },
{ "@type": "AdministrativeArea", "name": "Cowlitz County, WA" }
],
"hasOfferCatalog": {
"@type": "OfferCatalog",
"itemListElement": [
{ "@type": "Service", "name": "Drywall Installation" },
{ "@type": "Service", "name": "Drywall Repair" },
{ "@type": "Service", "name": "Hand Texture" }
]
}
} That's how "popcorn ceiling removal" and "decorative plaster" landed in Google's answer earlier. They weren't buried halfway down a paragraph. They were declared by name, in a list a model can read without guessing, with the service area stated as Kalama and Cowlitz County right next to them.
This is the part that lives in code, not in your copy, which is why a lot of template-built sites never get it right. It's baked into every site we build, because a fast site with clean structured data is the floor for showing up in AI search.
Be a Business the Model Already Trusts
Models don't just read your site. They cross-check your business against the rest of the web before they vouch for you in an answer. The more consistent, credible places you show up, the more citable you become.
So get the off-site house in order. A complete Google Business Profile. Accurate listings on the directories that matter for your industry, not a hundred junk ones. Reviews you earned honestly. A mention or two in local press if you can get it. And an About page with real people, real names, and real credentials, because that's what experience and trust look like to a model deciding if you're the real thing.
This is the same entity-building work that wins home services, healthcare, and legal businesses in regular local search. AI search just raised the stakes on getting it consistent.
Rather have us handle all of this?
We build the fast site, write the quotable content, and wire up the schema so you get found. Reach out for a free quote.
How to Know If It's Working
You can't open a dashboard and see your "ChatGPT ranking." The data just isn't exposed the way Google Search Console exposes search. So you measure it three honest ways:
- Re-run your prompt sheet monthly. Same questions, same platforms, every month. Watch for the month your name starts appearing and which competitors drop off.
- Watch your analytics for AI referrals. In GA4, look for traffic from sources like chatgpt.com, perplexity.ai, and gemini.google.com. The volume is still small, but it's some of the highest-intent traffic you'll get, because the person already got a recommendation before they clicked.
- Track your brand mentions. Set a recurring search for your business name and watch where new mentions show up. More mentions on trusted sources usually shows up later as more citations.
And be skeptical of precision. If an agency hands you a slick "AI search ranking report" with exact positions and percentages, ask where the numbers come from. Most of that is estimated. The honest answer right now is directional: are you getting named more often than you were three months ago, yes or no.
Wherever your business is, if you want to see where you currently stand, we'll run your real customer questions through the major AI tools and show you exactly where you show up or don't. No commitment, no pitch. Just reach out and we'll send you the rundown.
We Do This Nationwide
Two of the businesses above are in our backyard near Vancouver, Washington. One is in San Francisco. That's the point. The playbook doesn't care where you are. We run SEO and AI search for businesses all over the country, and the mechanics are the same in every market: a fast static site, content that answers real questions, and schema that spells out who you are and where you work.
A few of the markets we work in:
That's a sample, not the whole map. See the full list of locations we serve, or just tell us your city and we'll take it from there.
Getting Cited by AI, Answered
Sources & References
- Aggarwal et al. "GEO: Generative Engine Optimization" (KDD 2024, Princeton / Georgia Tech / Allen Institute / IIT Delhi)
- Surfer SEO "AI Overviews Study" (Dec 2025)
- Google "Introducing AI Mode in Search" (2025)
- Google Search Central "Intro to Structured Data Markup"
- Search Engine Land "Mastering Generative Engine Optimization in 2026: Full Guide"