If you’ve spent any time in local SEO, you already know the standard answer to “how does Google Maps rank businesses”: relevance, distance, prominence. It’s not wrong. It’s just not nearly the whole picture, and recent research into Google’s local search infrastructure makes that gap obvious.

In late 2025, independent researchers published what’s likely the most detailed public analysis of Google’s local search systems to date, built from a mix of decompiled binaries, network traffic analysis, and cross-referencing with the 2024 Google Search API leak. The findings, reported by Search Engine Land, describe roughly 72 internal ranking signals tied to a system called Geostore, Google’s internal representation of the physical world, along with hundreds of data providers, a separate ranking layer called Oyster Rank, and a rendering system called Mapcore that decides what actually shows up on the map.

None of this has been confirmed by Google as an active, complete ranking algorithm. It’s reported internal architecture, not an official disclosure. But even treated cautiously, it tells you something the relevance-distance-prominence model never could: Google Maps isn’t running one ranking formula. It’s running a pipeline, and your business has to clear every stage of it, not just the ones you’ve been optimizing for.

That’s the shift this article is about. Not “here are 72 things to add to your checklist,” but a genuine change in what local SEO work needs to look like.

Why the traditional Local SEO checklist is no longer enough

For most of the last decade, local SEO practice has centered on a fairly stable set of activities: filling out a Google Business Profile completely, picking the right primary and secondary categories, collecting reviews, uploading photos, keeping name-address-phone (NAP) data consistent across directories, listing accurate hours, and building citations. Proximity to the searcher and general prominence rounded out the picture.

This checklist earned its place because it worked, and because Google gave SEOs just enough information (the relevance/distance/prominence framing) to make it feel complete. It still matters. A business with a wrong category, inconsistent NAP data, or a thin profile is not going to compete, no matter how sophisticated the rest of its strategy is.

What’s changed is the ceiling. The checklist gets a business into contention. It doesn’t explain why two businesses with equally strong profiles rank differently, why some businesses show up for adjacent, unoptimized queries, or why a business three miles away sometimes beats one two blocks from the searcher. Those outcomes come from further down the pipeline: how well Google’s systems understand the business as an entity, how it’s matched against the specific query, and how it’s scored against everything else in the running.

Local SEO isn’t replacing the checklist. It’s absorbing it into something larger.

The difference between a business listing and a business entity

A listing is what a user sees: a name, an address, a phone number, a star rating, some photos. An entity is Google’s internal model of the real business behind that listing, built by pulling together and reconciling information from many independent sources.

The research reported by Search Engine Land describes nearly 800 data source providers feeding into Geostore, with built-in mechanisms for provenance, trust scoring, priority, and conflict resolution when sources disagree. Practically, that means Google isn’t just reading your Google Business Profile. It’s cross-checking it against your website, structured data, directory listings, review platforms, press mentions, and even the Knowledge Graph, through a connector reported as Webref that links web documents to entities via their Knowledge Graph identifier.

Here’s a concrete example. A dental practice lists itself as “Smith Family Dentistry” on its Google Business Profile, but its website footer and schema markup say “Smith Dental Group.” A local directory has it as “Dr. John Smith, DDS.” None of these are wrong, exactly. But from an entity-resolution standpoint, Google now has to decide whether these are the same business, three different businesses, or something in between. Every inconsistency adds friction to that decision. Every one you remove makes the entity clearer, and a clear entity is what gets confidently matched to search queries.

This is why “NAP consistency” undersells what’s actually needed. It’s not just matching a name and phone number across ten directories. It’s making sure every representation of the business, on every platform Google can see, resolves to a single, unambiguous entity.

What the reported 72 Google Maps ranking signals actually mean

The categories of signals described in the research include reviews, web query volume, listing impressions and profile opens, direction requests, website click-throughs, general popularity, prominence, brand or chain membership, and even road usage data. It’s a legitimately interesting list, and it’s worth understanding at the category level. It is not a checklist to implement in order.

Here’s the distinction that matters. The existence of an internal signal name doesn’t tell you whether that signal is currently active for your query type, how much weight it carries relative to the others, whether it affects candidate retrieval or final ranking (those are different stages), or whether it applies uniformly across categories like restaurants, law firms, and auto repair shops. Reported internal signal names are exactly that: reported and internal. Confirmed ranking factors are the small set Google has actually acknowledged, largely still relevance, distance, and prominence at the public-facing level.

What the 72-signal list is genuinely useful for is understanding what Google’s systems appear to care about as categories of evidence:

Signal category	What it likely reflects	Practical takeaway
Reviews (volume, recency, content)	Customer-verified quality and relevance	Reviews are entity information, not just a trust badge
Search demand and query volume	How often people look for this type of business	Category and service accuracy matter more in high-demand niches
Listing engagement (opens, clicks, direction requests)	Real user interest translated into behavioral data	A complete, compelling profile earns its own ranking signal through usage
Website interactions	Confirmation that the entity has a substantive web presence	Local SEO and website SEO are no longer separable
Chain or brand relationships	How a location connects to a parent brand	Multi-location businesses need explicit brand-to-location linking
Geographic and landmark relevance	Confidence about where a business actually operates	Address accuracy and area descriptions still carry real weight

Treat this as a map of what Google’s systems are trying to learn about your business, not a scorecard to fill in.

Why Google Maps may not have one single ranking algorithm

The research describes something closer to a pipeline than a formula, and Search Engine Land’s reporting is explicit that the 72 signals are not “the algorithm.” Geostore has its own scoring layer (Oyster Rank). Search adds separate systems for query understanding, geography, and candidate generation on top of that. There’s reportedly a distinct scorer that runs on-device for personalization. And Mapcore, the rendering system, makes its own decisions about what actually gets drawn on the map, independent of ranking eligibility.

Based on that reporting, here’s an original framework for thinking about the stages a business likely has to clear:

Entity Understanding → Query Interpretation → Candidate Retrieval → Semantic Matching → Geographic Relevance → Quality and Importance Evaluation → Reranking → Personalization → Result Presentation → Map Visibility

Walking through it briefly: Google first has to understand what your business is (entity understanding), then figure out what the searcher actually wants (query interpretation), then pull a pool of plausible matches from its index (candidate retrieval), then check how well each candidate semantically fits the query (semantic matching), then weigh how geographically appropriate each one is for this specific search (geographic relevance), then score quality and importance signals like reviews and popularity, then apply reranking logic that may reorder results based on additional context, then personalize for the individual searcher where relevant, then decide how to present the result, and finally, separately, decide whether and how to render it on the map itself.

A business can be strong at one stage and weak at another. A well-reviewed, prominent business with a poorly described website might sail through quality evaluation but fail semantic matching for specific service queries. That’s the practical value of thinking in stages: it tells you where to actually look when something isn’t working, instead of defaulting to “we need more reviews.”

Rethinking proximity and geographic relevance

Proximity still matters, but “closest business wins” was always an oversimplification, and it’s worth retiring for good. Geographic relevance appears to be evaluated in context: what the query is, where the search is coming from, how dense the local business landscape is for that category, how many plausible candidates exist nearby, and how strongly a business is tied to a neighborhood or landmark.

In a dense urban core with fifteen competing coffee shops within four blocks, proximity differences of a few hundred meters may barely register against quality and relevance signals. In a sparse suburban or rural area, a business fifteen minutes away might be the most relevant result simply because nothing closer exists. Search intent shifts this further. Someone searching “emergency plumber” is likely to see a wider geographic net than someone searching “coffee shop,” because the system is inferring the searcher’s tolerance for distance based on category and urgency.

For businesses outside the tightest physical radius, the practical response is to lean harder on relevance and quality signals to compensate. That means clear service-area descriptions, neighborhood and landmark references where genuinely accurate, and content that answers the specific queries your target area is actually searching, rather than assuming proximity alone will carry you.

Why websites and local SEO are becoming increasingly connected

This might be the most consequential shift for SEO teams to internalize. Local SEO and organic website SEO have historically been run as separate workstreams, often by separate people, sometimes even separate agencies. The entity model makes that separation increasingly costly.

A website is one of the richest sources of entity information Google has access to. It’s where a business can explain, in its own words and structured data, exactly what it is, what it offers, and where it operates, with far more nuance than a Google Business Profile’s limited fields allow. Location pages, service pages, internal linking between a brand’s hub page and its individual locations, structured data marking up address and service information, and even external references and citations pointing back to the site all feed into how confidently Google resolves the business entity and matches it to relevant queries.

A business with a strong Google Business Profile but a thin, generic website is handing Google less to work with than a competitor whose website reinforces and expands on everything the profile says. Treat the website as the canonical source of truth for the entity, not a formality that exists alongside the “real” local SEO work happening on the profile.

Semantic Local SEO: moving beyond categories and keywords

Search engines increasingly appear to understand businesses through concepts and relationships, not just exact-match categories and keywords. A restaurant’s Google Business Profile category might simply say “Italian restaurant,” but the entity Google builds around that business can plausibly include vegetarian options, specific dishes, family-friendliness, outdoor seating, noise level, and service speed, drawn from reviews, website content, photos, and structured data.

This is what “semantic local SEO” means in practice: making sure the concepts that actually differentiate your business are stated somewhere Google can read them, in natural language, without resorting to keyword stuffing. If your restaurant genuinely has a strong vegetarian menu, say so clearly on the site and in a way that shows up in photos and reviews too. If your dental practice specializes in pediatric care, don’t just tuck “pediatric dentistry” into a category field, describe the actual experience of bringing a child in.

The failure mode to avoid is manufacturing attributes you don’t have to chase semantic coverage. Overstating “family-friendly” or “quiet atmosphere” when it isn’t true creates a mismatch between what Google’s systems infer and what customers actually experience, which shows up in reviews and behavioral signals anyway.

Reviews as a source of entity information

Reviews do more work than most businesses give them credit for. Beyond the aggregate star rating, the actual text of reviews is a source of entity-level information: which products and services customers mention, what themes recur (fast service, friendly staff, long waits), and what specific attributes show up unprompted.

The ethical path here is straightforward: make it easy for real customers to leave reviews, respond to them genuinely, and never incentivize positive reviews, fabricate them, or coach customers to include specific keywords. Beyond being against platform policy, manufactured reviews tend to read as generic and don’t reinforce the specific, differentiating entity signals that authentic reviews naturally provide. A steady stream of honest reviews that happen to mention your specialties, your neighborhood, and your service quality will do more for entity clarity than any amount of engineered language ever could.

What multi-location businesses should learn from entity-based local SEO

Franchises, retail chains, healthcare groups, hotel brands, and multi-location service businesses face a harder version of the entity problem: they need Google to understand both the parent brand and each individual location as connected but distinct entities.

The chain here matters: parent brand → individual location → website location page → Google Business Profile → external references. Every link in that chain needs to be explicit and consistent. The parent brand’s website should clearly link to and identify each location. Each location needs its own Google Business Profile, its own accurate address and service data, and ideally its own genuinely unique location page, not a templated page with the city name swapped in.

Duplicate or thin location pages are a common failure point. A page that’s 90% identical boilerplate across fifty locations, with only the address changed, gives Google very little to differentiate one location from another and can actively work against ranking for location-specific queries. Real local relevance, staff bios, location-specific service notes, genuine local reviews, area-specific content, does more to establish each location as its own credible entity than volume of pages ever will.

Ranking versus visibility on the map

These are not the same thing, and the research is a useful reminder why. Being retrieved as a candidate, ranking well within local search results, appearing in the local pack, and being visibly rendered as a label on the map itself all appear to be handled by at least partially separate systems. The reported Mapcore rendering layer, with tens of thousands of distinct map and label styles, makes its own decisions about what’s drawn and how, independent of whether a business is ranking well in search results.

This matters practically because grid-based rank tracking tools and map screenshots, while useful, are measuring visibility at a specific moment and zoom level, not the complete picture of how a business is performing across every stage of retrieval and ranking. A business can be performing well in actual search result rankings while looking inconsistent in map-label visibility, or vice versa. Don’t treat either measurement as the full story.

AI and the future of local search

Conversational and AI-powered search is pushing local queries toward genuine natural language, and that raises the bar on everything discussed above. Consider a query like “find a restaurant nearby for six people with vegetarian options, good service reviews, and a short wait.” Answering that well requires understanding the entity (is this actually a restaurant that can seat six), its attributes (vegetarian options), its reputation (service quality, drawn from reviews), its current context (wait times), and its geography, all at once, then reasoning across all of it to produce a single recommendation.

This is precisely why entity clarity, semantic coverage, and authentic reviews compound in value as AI Overviews, Gemini, and third-party tools like ChatGPT and Perplexity increasingly mediate local decisions. These systems draw on many of the same underlying signals, website content, structured data, reviews, entity consistency, that traditional local search ranking depends on. Businesses that have already done the entity-clarity work described in this article are better positioned to be understood and recommended by AI systems, not just ranked in a traditional local pack. This is the practical overlap between local SEO and generative engine optimization: the same groundwork serves both.

How businesses should adapt their Local SEO strategy

1. Strengthen entity consistency

Audit how your business name, address, phone number, category, and description appear across your Google Business Profile, website, directories, and any platform Google is likely to crawl. Resolve every inconsistency, not just the obvious ones.

2. Improve website-to-entity clarity

Make sure your website answers, in plain language and structured data, who the business is, what it offers, where it operates, and which locations belong to it if you have more than one.

3. Create useful and unique location pages

Replace templated, near-duplicate location pages with pages that include genuinely local information: specific services offered at that location, local staff, area-specific content, and real local reviews.

4. Improve semantic coverage

Clearly describe your actual services, products, specialties, and the specific customer needs you address, in natural language across your website and profile, without stuffing keywords or inflating claims.

5. Encourage authentic customer feedback

Build review generation into your normal customer experience. Never incentivize, fabricate, or script reviews. Real feedback carries entity information that manufactured feedback can’t replicate.

6. Connect parent brands and individual locations

For multi-location businesses, make the brand-to-location relationship explicit everywhere: website architecture, Google Business Profile chain settings, and structured data.

7. Use structured data appropriately

Implement LocalBusiness and related schema accurately to support clear machine understanding of your entity. Don’t treat schema as a ranking shortcut. It supports understanding; it doesn’t replace the underlying content.

8. Monitor entity accuracy beyond Google Business Profile

Periodically check how your business is represented in search results, on your website, in directories, on the map, and in AI-generated answers. Entity drift happens quietly across dozens of sources.

9. Optimize for questions, not just keywords

Identify the specific, often multi-part questions customers are asking before choosing a local business, and make sure your content answers them directly and concisely.

10. Treat local SEO as part of a broader SEO, AEO, and GEO strategy

Local SEO, organic SEO, answer engine optimization, and generative engine optimization increasingly draw on the same underlying entity and content signals. Plan them together rather than as separate workstreams with separate owners.

Key Takeaways

• Google Maps ranking appears to run through a multi-stage pipeline (entity understanding, query interpretation, retrieval, matching, geographic relevance, quality evaluation, reranking, personalization, presentation, and map rendering), not a single formula.
• The reported 72 ranking signals are categories of evidence Google’s systems may weigh, not a checklist to implement in order or a confirmed active algorithm.
• Local SEO is shifting from managing a Google Business Profile to managing a consistent, well-documented business entity across the entire web.
• Websites are now core local SEO infrastructure, not a separate discipline from Google Business Profile optimization.
• Proximity is contextual, shaped by query type, competitive density, and search intent, not a simple closest-business-wins rule.
• Multi-location businesses need explicit, consistent connections between parent brand and individual locations, backed by genuinely unique location content.
• Reviews function as entity information, not just a trust score, so authenticity matters more than volume alone.
• Ranking eligibility, search ranking, local pack inclusion, and map visibility are related but distinct outcomes.
• The entity-clarity work that strengthens local search also strengthens visibility in AI Overviews and conversational search tools.

Frequently Asked Questions

What are the Google Maps ranking factors?

Google has officially confirmed relevance, distance, and prominence as the core factors. Recent research reports roughly 72 additional internal signals within Google’s Geostore system, covering categories like reviews, search demand, listing engagement, and brand relationships, though these are reported findings rather than a confirmed, complete algorithm.

Is the “72 ranking signals” research confirmed by Google?

No. It comes from independent research based on decompiled binaries, network analysis, and the 2024 Google Search API leak, reported by Search Engine Land. Google has not officially confirmed these as active, weighted ranking factors.

What is entity SEO in the context of local search?

Entity SEO is the practice of making sure Google can accurately and consistently understand your business as a single real-world entity across every source it references, including your Google Business Profile, website, directories, and reviews, rather than optimizing each of those in isolation.

Does proximity still matter for Google Maps ranking?

Yes, but it’s evaluated in context rather than as an absolute rule. Query type, local competitive density, and search intent all influence how much weight proximity carries in a given search.

How does a business website affect Google Maps rankings?

A website provides detailed information that helps Google understand and confirm the business entity, including services, locations, and structured data. Strong website content supports both organic and local search performance, since the two increasingly rely on the same underlying signals.

What should multi-location businesses do differently for local SEO?

Explicitly connect the parent brand to each individual location across the website, Google Business Profile, and structured data, and build genuinely unique content for each location rather than duplicating a template across all of them.

How should businesses prepare for AI-powered local search?

Focus on the same fundamentals that support traditional local search: clear entity information, authentic reviews, and detailed, accurate website content. AI Overviews and conversational search tools draw on many of the same signals used in local search ranking.

Is Google Business Profile optimization still important?

Yes. A complete, accurate profile remains necessary for a business to be considered at all. It’s no longer sufficient on its own, since ranking depends on stages beyond the profile itself.