The Patent Is Public. Most Service Business Owners Have Not Read It.
Search Engine Land published Google's LLM patent in detail, and the implications are significant for every service business running on local search visibility. Google is not just ranking pages anymore. It is building structured knowledge profiles, what the patent calls entity representations, from your website content, your reviews, your business listings, and any public data it can find about your operation. Those profiles power AI Overviews. If your entity profile is incomplete, inconsistent, or thin, you are invisible inside the AI result, and that is where 40% of queries now end.
This is not a future problem. It is a present one. The shift happened faster than most operators noticed, and the businesses that figure this out in the next six months will hold positions that are difficult to displace.
What Google's Patent Actually Says
The patent describes a system for building entity-level knowledge representations by aggregating signals across sources. Google is not looking at your website in isolation. It is cross-referencing your website with your Google Business Profile, your review text, citations in directories, mentions in local news, and any structured data it can extract. It is building a profile of who you are, what you do, who you serve, and whether those signals agree with each other.
The key word in the patent is consistency. When your website says you serve commercial clients in the Phoenix metro, your GBP says "Phoenix, AZ," your reviews mention commercial buildings and office complexes, and your citations reference the same service area, Google can build a coherent entity profile. When those signals contradict each other, or when some of them are missing entirely, the profile is weak.
A weak entity profile means Google's AI has low confidence in what your business is. Low confidence means you do not appear in AI Overviews for the queries that matter. You may still rank in traditional blue-link results, but you are absent from the section where 40% of the search behavior is now concentrated.
The 40% Number Deserves More Attention
Bain published data showing that 40% of queries now stay inside AI-generated results. That figure does not mean 40% of all internet searches. It means 40% of queries where an AI Overview is triggered, which skews heavily toward informational and transactional queries in services, home improvement, legal, medical, financial, and professional categories.
Those are exactly the categories where service businesses live.
During my time running analytics at Hartford and Munich Re, we had a phrase in the underwriting operations: "the risk you do not see is the risk that kills you." We were talking about catastrophic loss exposure, but the principle holds here. Most service business owners are watching their traditional organic rankings and seeing stability. They are not watching AI Overview impressions because Google Search Console has only recently started surfacing that data meaningfully. The business is eroding in a channel they are not monitoring.
This is the casualty drill moment. You do not wait until you see the revenue drop to start tracking AI Overview visibility. You build the entity footprint now, while there is still time to establish position.
The ATLAS Model Applied to Entity SEO
The ATLAS Model is a five-part framework for building and auditing your digital asset footprint in a way that compounds over time. ATLAS stands for Authority, Trust signals, Listings accuracy, Audience specificity, and Structured data. Applied to entity SEO for AI Overviews, each layer has a specific job.
Authority is your website's topical depth. Google's entity system rewards businesses that demonstrate expertise across the full range of their service. If you are a commercial HVAC contractor, your site should cover commercial HVAC maintenance, commercial HVAC installation, commercial HVAC emergencies, and the specific building types you serve. Not surface-level pages. Substantive content that reflects how a commercial HVAC expert actually thinks. The AI is reading for domain knowledge, not keyword density.
Trust signals are your reviews. Not just the star rating. The text. Google's patent specifically describes extracting semantic information from review content to build entity profiles. When your reviewers use the same language to describe your service that you use on your website, that is a consistency signal. When reviews mention your name, your service type, your location, and a specific outcome, they are contributing structured entity data without you doing anything. This means review strategy is now an SEO strategy.
Listings accuracy is table stakes, but most businesses fail here. Your business name, address, phone number, and service description must be identical across every directory, citation, and data aggregator. Not similar. Identical. The entity system is pattern-matching across sources. Inconsistency registers as ambiguity, and the AI does not include ambiguous entities in AI Overviews.
Audience specificity means your content, your GBP description, and your review responses must speak to a defined customer type. "We serve all businesses" is a weak entity signal. "We serve multi-unit commercial properties, property management firms, and institutional real estate owners in the Phoenix metro" is a strong one. Specificity helps the AI understand where to surface you.
Structured data is the explicit layer. Schema markup for your business type, services, service area, reviews, and FAQs gives Google machine-readable confirmation of what your entity is. It is the fastest way to strengthen a weak entity profile because it translates your existing content into the format the AI is already looking for.
Sovereignty Stack for Search Visibility
Here is the doctrine point: you do not own your Google rankings. You never did. What you can own is the quality and consistency of your entity footprint, and that footprint exists across assets you control: your website, your review responses, your GBP description, your schema markup, your content calendar.
The Sovereignty Stack principle says that your long-term visibility depends on assets you control, not on any single platform's algorithm. Entity SEO aligns with that principle better than traditional keyword SEO ever did. Traditional SEO optimized for a ranking signal that Google could change overnight. Entity SEO builds a knowledge profile that compounds across every platform where your business appears. That profile becomes an asset on your balance sheet, not just a tactic in your marketing plan.
When a private equity buyer looks at a service business today, they are increasingly asking about organic search sustainability. A business with a strong entity footprint and consistent AI Overview visibility has a defensible customer acquisition channel. A business that depends on paid search or a thin organic presence has a channel risk that shows up in the multiple. Build the entity footprint as if it is part of your exit thesis, because it is.
What the Playbook Looks Like in Practice
The 90-day entity audit for a service business runs in three phases.
In the first 30 days, you audit consistency. Pull every citation, directory listing, and GBP field. Build a master record for your business name, address, phone, service description, and service area. Fix every inconsistency. This is unglamorous work. It is also the foundation of everything else.
In days 31 through 60, you build topical depth. Identify the five to eight service and audience categories that define your business. Write substantive pages for each. These are not blog posts. They are reference-level pages that reflect genuine expertise. Link them together. Add schema markup to every page.
In days 61 through 90, you systematize the trust signal layer. Create a review request process that asks customers to mention specific services and outcomes. Write GBP review responses that reinforce your entity signals. Begin a content calendar for monthly additions to your topical depth. At day 90, run a Google Search Console audit for AI Overview impressions. Compare to your baseline. The entity footprint you built in 90 days will outperform what most competitors have built in years.
Internal Resources
For operators running service businesses who want to build a complete digital asset strategy, see the Sovereignty Stack breakdown for service businesses and the 90-Day Bottleneck Audit guide.
Doctrine Connection
The shift to AI Overviews is not a search engine update. It is a balance sheet event. Your entity footprint is an asset. If it is weak, inconsistent, or thin, you are carrying an invisible liability in your customer acquisition channel. If it is strong, specific, and consistent, you hold a position that compounds over time and that acquirers will recognize as defensible. Build it like an asset, because the operators who treat it as a tactic will be replaced by the ones who treat it as doctrine.
FAQ
What is an entity profile in Google's context? It is a structured knowledge representation Google builds for your business by aggregating signals from your website, Google Business Profile, reviews, citations, and public data. This profile determines whether and how your business appears in AI Overviews.
Do I need to be ranking on page one to appear in AI Overviews? Not necessarily. AI Overviews draw from entity profiles, not just traditional ranking positions. A business with a strong, consistent entity footprint can appear in AI Overviews even without ranking in the top three organic positions. However, a strong entity profile and strong traditional rankings are correlated because both reward the same underlying signals.
How long does it take to build a strong entity footprint? A disciplined 90-day effort can produce meaningful improvements in AI Overview visibility, particularly for service businesses in mid-competition markets. High-competition markets take longer, but the compounding nature of entity SEO means that consistent effort over 6-12 months produces positions that are difficult for competitors to displace.
What is the biggest mistake service businesses make with their GBP? Inconsistency between the GBP and the website. Specifically: service descriptions that do not match the website copy, service areas defined differently across platforms, and business names that vary by punctuation or abbreviation. These seem like minor issues. To the entity system, they register as ambiguity and suppress AI Overview visibility.
Is this relevant to businesses that get most of their leads from referrals? Yes. Even referral-driven businesses have prospects who validate them through search before calling. If an AI Overview returns a competitor's entity profile when someone searches your business type in your market, you lose validation opportunities even when someone was already inclined to call you. Entity SEO is reputation infrastructure, not just acquisition infrastructure.