How AI systems find and recommend local businesses (part 1 of 2)

On September 20, we shared a BrightLocal finding in an earlier post: 45% of consumers now use AI tools to find local business recommendations, up from 6% the year before.¹ This post adds more data on the advantage local businesses gain when they optimize their sites for answer engines, and it explores how AI systems choose which businesses to recommend.

The terms we use

• AI and answer engines: we use the two interchangeably to describe any large language model (LLM) chatbot, such as ChatGPT or Gemini; voice assistants, such as Alexa; and AI-generated summaries, such as Google’s AI Overviews.
• AI Overview: Google’s AI-generated answer, placed at the top of traditional search results.
• Zero-click search: a search that ends without the user clicking through to any website.
• SEO: search engine optimization.
• AEO: answer engine optimization.

ChatGPT, Claude, Gemini, Perplexity, and Google’s AI Overviews draw on an expanding evidence ecosystem to recommend local businesses, and more consumers act on those recommendations every year. That creates an opportunity for the local businesses that optimize for these systems. How the major AI ecosystems collect, rank, and suggest local businesses tells a business what it must do to raise its odds of being cited.

Why AI search results matter for small businesses

AI-based searches for local businesses are still a small share of the searches run each day, but they are growing, and they convert at higher rates than traditional non-branded Google searches.² Local businesses that optimize their online presence for answer engines build an advantage over their local competitors and over national chains.

ChatGPT and Gemini each run their own retrieval and also pull data from other providers, such as map data, review sites, and business directories.

For two decades, businesses paid to appear in the prime real estate at the top of the search results. In a 2005 eye-tracking study by Enquiro, Did-it, and Eyetools, all 50 participants viewed the top three organic results,³ and the study’s co-author later reported that more than 80% of first clicks came from the top four listings.⁴ Today, AI Overviews take that spot for most local searches. A 2025 Whitespark study of 540 local business-type queries found AI Overviews on 68% of them.⁵

As a small business, you are either part of the answer, or your competitors are.

What AI local recommendations are not

AI systems do not take Google’s rankings for local businesses and repackage them as answers. Each AI product gathers and weighs its own evidence. In a 2026 Steady Demand study of 14,472 citations, ChatGPT and Gemini cited the same domains only 8% of the time.⁶ Understanding those approaches, and acting on them, gives a local business a competitive advantage.

Optimizing your site for AI visibility is not about keywords or long-tail phrases. It is about creating touchpoints across multiple platforms that lead back to your business. AI visibility does not come from rankings; it comes from setting up and maintaining the right signals so AI systems can see the patterns that connect your online presence.

AI visibility is not a ranking. It is a pattern-recognition problem.

The core model: AI looks for patterns and corroboration

AI systems don’t read, at least not the way people do. They search for patterns across multiple sources, such as websites, social media, business directories, and consumer reviews, to build a picture of a local business. Each system learns from a large set of training data before anyone asks it a question. It also uses real-time retrieval, a live web search at the moment of the question, to add current information. Together, the training data and the live search produce a consensus view of your company.

Anthropic’s documentation for its AI assistant, Claude, states when each one applies. Claude answers directly from its training data when a question draws on stable knowledge, such as established facts. It searches the web when the answer depends on information that is current or changing, such as prices or details about specific organizations that might have changed.⁷

AI doesn’t rank your website or business in the traditional sense. It reads what Google, the directories, and your own website say about you, and it recommends the businesses those sources agree on.

ChatGPT: from Bing dependency to multi-source local retrieval

From Bing to its own retrieval stack

In May 2023, Microsoft announced Bing as the default search experience in ChatGPT.⁸ At the time, the model behind ChatGPT had a September 2021 training cutoff.⁹ In October 2024, OpenAI launched ChatGPT search, built on a fine-tuned version of GPT-4o and drawing on third-party search providers and content partners.¹⁰ By 2026, Peec AI research found ChatGPT running its own retrieval index, internally called Labrador, and blending it with Google, Bing, and at least eight outside providers. Yelp and TripAdvisor supply local data as licensed partners.¹¹¹²

These approaches to indexing sites and corroborating evidence across partner services will keep changing as the industry matures. Each change gives small businesses a new evidence ecosystem to tap into and optimize for.

Why aggregators outweigh your website (for ChatGPT)

In Steady Demand’s 2026 study, ChatGPT cited a business’s own website only 15.9% of the time and relied more on Reddit and business directories.¹³ BrightLocal’s analysis of 1.9 million AI citations found Yelp in 80% of ChatGPT’s local answers.¹⁴

That gives local businesses a clear path for answer engine optimization, and it is not as simple as adding keywords and phrases to a site. Site content needs to be relevant and fresh. Interactions between brands and consumers need to be genuine and span multiple platforms. Online reviews matter, and not just the reviews on your Google Business Profile.

Gemini: Google Search and Maps still dominate

Google’s infrastructure still dominates local business search, but not the way it did three or five years ago. Search for a local business on Google, and 68% of the time an AI Overview sits at the top of the results.⁵

In early 2026, fewer than a third of U.S. Google searches ended with a click. SparkToro’s analysis of searches from January through April 2026 put the zero-click rate at 68%, up from 60% in 2024. Those searches never left Google’s ecosystem for the open web. AI Overviews now appear on more than 20% of Google searches, and when they do, click-through rates drop by nearly 60%.¹⁵ Click-through rate is the share of searchers who click a result.

Impacts of zero-click on local business

The rise of zero-click searches means local businesses that AI does not cite carry a competitive burden they might not yet realize. Whether it is an AI summary, a featured snippet, a knowledge panel, or a local map pack, users who never leave the platform they searched from never visit a business’s website. For a restaurant, the customer’s final action online is no longer ordering a meal or making a reservation on its website. It is tapping “Order online” or “Call,” tapping “Directions” to open a map app, or clicking “Reservations” in the search results.

To stay relevant, local businesses that have chased traffic and clicks for their websites need to pivot.

Grounded in Google’s own data

Google’s AI assistant, Gemini, is grounded in real-time Google Search results: its answers draw on live web content.¹⁶ For geographic questions like “ice cream shops near me,” Gemini can also pull place data from Google Maps, which covers more than 250 million places.¹⁷ That place data includes the business details and ratings from Google Business Profiles. Google publishes no formula for how much ratings count. These data sources show why a business needs a visible pattern online for AI systems to find, recognize, and ultimately cite as an answer.

Your website is the main citation source, so local SEO still matters

Compared to ChatGPT, Gemini leans on business websites. In Steady Demand’s August 2026 study, 59.9% of Gemini’s local citations pointed to a business’s own website, more than directories, review platforms, and forums combined.⁶ A September rerun put that share at 49.6%, still the largest single source.¹⁸ Local SEO still matters, and the rise of AI-based search does not discount it. SEO and AEO are complementary; giving up one for the other is short-sighted and ultimately harmful for your brand.

Whitespark’s 2026 expert survey found that 3 of the top 5 AI search visibility factors are citation factors.¹⁹ In local search, a citation is a mention of your business’s name, address, and phone number on another website, such as a directory. In Whitespark’s words, “In AI SEO, mentions (citations) are the new link.”

For Gemini, your website still matters. For ChatGPT, your presence around the web may matter more.

Part 2 of this series continues the breakdown.