SEO & Marketing

Jun 22, 2026 · 9 min read

Google Patented a New Way AI Understands Your Business — What Small Businesses Need to Do

On June 22, 2026, Search Engine Land reported on a newly published Google patent that reveals how the company's large language model (LLM) builds understanding of businesses, brands, and entities directly from web content. The patent describes a system where Google's AI reads your website, your Google Business Profile, your reviews, and third-party mentions — then constructs a structured "entity profile" that determines how (and whether) your business appears in AI Overviews, AI Mode, and conversational search results. For small business owners in Southern California, this changes what SEO means. The old playbook — stuff keywords into title tags and build backlinks — is no longer sufficient. The new playbook is about teaching Google's AI who you are, what you do, and why you matter. Here is what the patent actually says, what it means for your business, and the specific steps to take.

June 22, 2026 Category: SEO & Marketing 9 min read

What the Patent Actually Describes

The patent, analyzed by SEO researcher Rich Sanger at Search Engine Land, details how Google's LLM-based search system extracts entity information from web documents and builds structured profiles for businesses. Here is how it works in plain language:

Google's AI reads content across the web — your website pages, your Google Business Profile listing, reviews on Yelp and Google, social media profiles, news articles that mention you, and directory listings. From all of these sources, it extracts what the patent calls "entity signals": your business name, location, services offered, pricing, hours, customer sentiment, and areas of expertise.

These signals are then aggregated into a unified entity profile — essentially, Google's AI builds a dossier on your business. The completeness and consistency of this profile directly affects whether your business appears when someone asks Google's AI a question related to your services.

This is not theoretical. The patent describes the technical architecture behind features that are already live: AI Overviews, AI Mode, and Google's Search Agents. When someone asks "best taco shop in San Diego" and Google's AI returns a specific recommendation with your business name, it is pulling from this entity profile — not from a traditional keyword ranking.

Why This Matters More Than Any Core Update

Google core updates come and go. Some shake local rankings for a few weeks, then things stabilize. This patent is different because it describes a fundamental shift in how Google's search engine understands your business.

Traditional SEO works by matching keywords in your content to keywords in search queries. You optimize a page for "plumber Anaheim" and hope Google puts you on page one. The ranking signals are links, content quality, and on-page factors.

What the patent reveals is that Google's AI is running a parallel system on top of traditional search. This system does not match keywords. It matches entities — the actual identity, attributes, and reputation of your business — to the intent behind a user's question.

Think of it this way: a customer asks Google, "Who is the most reliable plumber in Anaheim with good reviews and fair pricing?" Traditional keyword SEO might rank you for "plumber Anaheim." But the entity-based system evaluates whether your profile matches the full question — reliability, reviews, and pricing — and decides whether to recommend you in the AI answer.

For small businesses, this means two things: First, you need to be visible in both systems (traditional search AND AI entity profiles). Second, the entity system is becoming more influential as AI search grows. We covered in our Search Console AI reports post that Google now lets you track AI Overviews and AI Mode impressions separately. That reporting exists because these are genuinely separate systems.

The Three Pillars of AI Identity SEO

Based on what the patent describes and what we know about how AI Overviews and AI Mode currently work, there are three pillars that determine how Google's AI understands your business:

Pillar 1: Structured Entity Data (Schema Markup)

Schema markup is the most direct way to feed entity signals to Google's AI. When you add LocalBusiness, FAQPage, Service, or Product structured data to your website, you are literally giving Google's AI a machine-readable profile of your business. The patent specifically references structured data as a key input for entity extraction.

If you have not added schema markup yet, this is your highest-priority action. We covered the step-by-step implementation in our schema markup guide. The basics: add LocalBusiness schema to your homepage with your name, address, phone, hours, services, and price range. Add FAQPage schema to any page that answers common customer questions.

Pillar 2: Consistent Business Identity Across the Web

The patent describes how Google's AI aggregates entity signals from multiple sources. This means consistency matters more than ever. If your business name is "Jose's Auto Repair" on your website, "Jose Auto Repair" on Google Business Profile, and "Jose Auto" on Yelp, Google's AI has to determine whether these are the same entity or three different businesses.

For small businesses in Southern California, this means auditing your NAP (Name, Address, Phone) across every platform where your business appears: your website, Google Business Profile, Yelp, Facebook, Instagram, BBB, industry directories, and any news articles or blog posts that mention you.

Make your business name, address, phone number, website URL, and service descriptions identical everywhere. Even small variations — "St." vs "Street", "&" vs "and" — can fragment your entity profile.

Pillar 3: Authority Signals and Sentiment

The patent describes how the AI evaluates not just what is said about a business, but the quality and sentiment of those signals. This maps directly to Google's E-E-A-T (Experience, Expertise, Authoritativeness, Trustworthiness) framework, which has been a ranking factor for years but is now being amplified by the entity-based AI system.

Practical implications for small businesses:

Reviews matter more than ever. Google's AI reads your star ratings, review text, and response patterns. A 4.8-star average with detailed, authentic reviews is a stronger entity signal than a 5.0 average with three generic reviews. Respond to reviews — both positive and negative — with specific, helpful replies.
Third-party mentions build entity authority. When a local newspaper, industry blog, or community website mentions your business, Google's AI picks that up as an authority signal. Getting featured in local media or contributing expert quotes to industry publications strengthens your entity profile.
Content depth establishes expertise. A website with 20 pages of shallow content sends weaker entity signals than one with 10 pages of detailed, specific content that answers real customer questions. The AI evaluates content quality, not just keyword presence.

What This Means for Different Types of Small Businesses

Local Service Businesses (Plumbers, Dentists, Roofers, Salons)

You are the biggest beneficiaries of this shift. Entity-based AI search favors businesses with clear, consistent profiles and strong review signals. Make sure your Google Business Profile is 100% complete with services listed, prices shown where possible, and fresh photos. Add schema markup to your website. Respond to every review. These actions directly feed the entity profile that determines AI recommendations.

Professional Services (Lawyers, Accountants, Consultants)

Authority signals carry extra weight for professional services. Google's AI evaluates expertise indicators — certifications, bar admissions, professional licenses, published articles, and speaking engagements. Make sure these are visible on your website with structured data. Create content that demonstrates deep expertise in your specific practice area, not generic industry content.

E-Commerce and Retail

Product schema becomes critical. Google's AI builds entity profiles for individual products and brands, not just businesses. Add Product structured data with pricing, availability, reviews, and specifications. Ensure your product descriptions are detailed and unique — the AI penalizes generic manufacturer descriptions duplicated across hundreds of sites.

Restaurants and Hospitality

Menu schema, review sentiment, and consistent NAP data across platforms drive your entity profile. Add Restaurant or Menu structured data. Keep your hours updated across all platforms (Google, Yelp, TripAdvisor). Post menus with prices so the AI can answer "how much is a burrito at [your restaurant]" directly.

How This Connects to Everything Else Happening in AI Search

This patent does not exist in isolation. It connects to several major developments from 2026:

Search Console Gen AI Reports (June 3, 2026): Google now shows you AI Overviews and AI Mode impression data separately. You can see which pages Google's AI is citing. The entity profile described in the patent is what drives those citations.
Google I/O 2026 (May 19, 2026): AI Mode became the default search experience for over 1 billion users. Search Agents now proactively surface information. All of these features rely on the entity extraction system described in the patent.
Lily Ray's AI Overviews Study (June 18, 2026): Analysis found that Google's AI cites self-serving listicles but recommends competitors 69% of the time. This is because the AI evaluates entity signals independently — it does not just parrot back what a single website claims. It cross-references your entity profile against the aggregate web consensus.
The May 2026 Core Update: This update explicitly affected AI Overviews rankings for the first time, confirming that Google's traditional and AI ranking systems are increasingly intertwined.

The pattern is clear: Google is building infrastructure to understand businesses as entities, not as collections of web pages. Every major search development in 2026 points in this direction.

What to Do This Week

Here is a prioritized action list for small business owners:

Add LocalBusiness schema to your homepage. Include your business name, address, phone, URL, hours, area served, services, and price range. Use Google's Rich Results Test to verify it works.
Audit your NAP consistency. Check your business name, address, and phone number on Google Business Profile, Yelp, Facebook, BBB, and any directory listings. Fix any discrepancies.
Add FAQPage schema to key pages. Identify the 5-10 questions customers ask most frequently. Create FAQ sections on relevant pages with those exact questions and detailed answers. Mark them up with FAQPage structured data.
Complete your Google Business Profile. Fill in every field: services with descriptions and prices, service areas, hours, attributes, photos, and the business description. An incomplete profile means an incomplete entity — and an incomplete entity gets fewer AI recommendations.
Respond to your Google reviews. If you have unanswered reviews, respond to them this week. Specific, helpful responses (not generic "Thank you!") strengthen the authority and engagement signals Google's AI reads.
Check your AI visibility. Open Google Search Console and look at the new Gen AI performance reports. See which pages are getting AI Overviews impressions and whether AI Mode is showing your business.

The Bottom Line

Google's patent confirms what the search industry has been theorizing: AI search does not work like traditional search. It does not match keywords to queries. It builds understanding of entities — real businesses with real attributes, reputations, and expertise — and matches those entities to user intent.

For small businesses, this is actually good news. The entity-based system rewards clarity, consistency, and genuine expertise. It rewards businesses that maintain accurate profiles, respond to customers, and create useful content. These are things any small business owner can do without enterprise SEO budgets.

The businesses that will struggle are the ones that ignore this shift — businesses with inconsistent listings across platforms, zero schema markup, unanswered reviews, and websites full of generic keyword-stuffed content. For them, AI search will be invisible.

The time to act is now. Google's AI search is not a future threat — it is the current reality for over a billion users. Your entity profile is being built whether you participate or not. The question is whether you shape it or let Google's AI fill in the blanks on its own.

If you want help optimizing your business for AI search visibility, implementing schema markup, or building a consistent entity presence across the web, reach out to PepeWebTech. We help small businesses across Southern California get found in the age of AI search.