The Google local 3-pack is the block of three business listings that appears above standard organic results for searches with local intent. Most users tap a result inside the 3-pack rather than scrolling to organic results below it, which makes those three positions the highest-value real estate in local search. The factors that decide which businesses fill those three slots aren’t secret, but they’re easy to misunderstand. The system isn’t a flat checklist; it’s a layered weighting that shifts by query type, business category, and searcher location. Reading the system correctly starts with Google’s own three-part definition and adds the operational realities that decide rankings in 2026.
The 3-pack is the only local result that matters at scale:
The local 3-pack is a SERP component Google shows for queries with local intent, displaying three business listings with map pin positions, ratings, and quick-action buttons (call, directions, website). The 3-pack appears above the standard organic results for queries Google identifies as local-intent, which covers most “near me” searches, branded local searches, and many category-only queries (“plumber,” “dentist,” “coffee shop”).
The user behavior on the 3-pack is concentrated. The top position typically captures the largest share of clicks and calls, the second position captures meaningful but lower volume, and the third position serves more as a comparison anchor than a destination. Positions four through seven (visible in the “more places” expansion) get a fraction of the engagement of the visible three.
This concentration is what makes 3-pack ranking the highest-value target in local SEO. A business that ranks #1 in the organic results below the 3-pack but doesn’t appear in the pack itself usually generates a small fraction of the call volume a 3-pack profile would. The visual prominence and action affordances (call button, directions, click-to-website) collapse the friction between intent and action in ways the blue-link results don’t match.
Google’s official triad: relevance, distance, prominence:
Google’s local ranking system is built on three factors the company describes explicitly in its own documentation. Relevance measures how well a business profile matches the searcher’s query. Distance measures how far each candidate business sits from the searcher (or from the location specified in the query). Prominence measures how well-known the business is, drawn from reviews, citations, links, and broader web signals.
The triad isn’t equally weighted across queries. Distance dominates “near me” queries where the searcher signals urgency about location: an HVAC search at midnight in July prioritizes proximity over reputation, since the closer technician will arrive faster. Relevance dominates specialty queries where the searcher cares less about proximity than about finding the specific service: someone searching for “personal injury attorney with truck accident experience” reads the category match before the distance. Prominence dominates competitive queries where multiple businesses sit at similar distances and similar relevance, so reputation breaks the tie: in a medspa-dense urban neighborhood, the practice with established review velocity and recognized clinicians ranks above newer competitors three blocks closer.
The Whitespark 2026 Local Search Ranking Factors Survey, based on input from 47 local SEO experts, breaks down the practitioner-observed weighting of signal groups: GBP signals carry the most influence, followed by review signals, on-page signals, link signals, behavioral signals, and citation signals. The specific percentages vary by methodology, but the rank order has stayed stable across multiple years of survey data.
What this means in practice: no single factor stands alone. A business with strong reviews but weak GBP completeness underperforms its competitors with weaker reviews and stronger profiles. A business with proximity advantages but thin prominence gets outranked by farther competitors with established reputations. The three pillars compound or compensate; treating one in isolation misses the mechanic.
Proximity is the lever you don’t control:
Distance is the factor businesses can least influence because it depends on the searcher’s location, not the business’s. A business in downtown Chicago ranks well for searches from downtown Chicago and progressively worse for searches from the suburbs as proximity falls off. The business address doesn’t change; the searcher’s location does.
The proximity calculation runs from the searcher’s device location (when available) or from the centroid of the location specified in the query (“dentist Chicago” calculates from Chicago’s geographic center rather than from any individual neighborhood). For queries without explicit location terms, Google uses the searcher’s IP location as the default reference.
The geo-grid rank tracking pattern shows how proximity shapes rankings at sub-city scale. Tools like Local Falcon and Places Scout sample rankings from a grid of locations across a service area (typically 7×7 or 9×9 for a metropolitan area), producing a heat map that shows where a business ranks well and where it loses position. The heat map almost always reveals proximity gradients: rankings strong near the business address, weakening as the sample points move away.
What proximity isn’t is a complete cap on visibility for businesses outside the immediate area. Strong relevance and prominence pull businesses into rankings from farther distances. The business that ranks at the edge of its effective proximity radius does so because the relevance and prominence signals are doing the work that distance can’t.
The strategic response is direct. Businesses with fixed locations work the variables they control (relevance, prominence, profile completeness) to expand their effective proximity radius rather than treating distance as the only constraint. The pillar where the business is furthest behind competitors is the lever to pull first.
Relevance is mostly about how the profile matches the query language:
Relevance signals come from how well the business profile and its connected content match the searcher’s words. The primary GBP category carries the heaviest single weight; the Whitespark survey identifies primary category as the most influential individual ranking factor. The category is the first filter Google uses to decide whether a business is even eligible to appear for a given query.
Secondary categories expand the relevance footprint. A primary “Dentist” with secondary “Pediatric dentist” and “Cosmetic dentist” appears for queries that match either secondary category in ways a Dentist-only profile doesn’t. Categories should reflect services performed rather than padding the field with adjacent categories that don’t match operations.
The service-list field within the GBP feeds relevance for query language match. A business listing “Class II composite restoration” matches that wording; one listing “Tooth-colored filling” matches the patient-facing wording instead. Most customers search the patient-facing term rather than the clinical one, which makes the customer-facing wording the more valuable surface.
Business name keywords are a contentious area. Names that legitimately include category keywords (“Chicago Plumbing Solutions”) get the relevance benefit; names stuffed with keywords for ranking purposes (“Best 24/7 Emergency Plumber Chicago LLC”) violate Google’s guidelines and risk suspension. The line between legitimate inclusion and stuffing isn’t always sharp, but profiles that drift past it draw enforcement attention.
Website content paired with the GBP contributes to relevance through the cross-reference Google does between profile and site. A profile for a dental practice with a thin website that doesn’t mention the services offered loses to a profile with a fully-developed site that confirms what the profile claims. The two surfaces validate each other.
Prominence is reputation that exists outside the profile:
Prominence aggregates signals from across the web about how well-known and trusted the business is. Review volume and rating, citation breadth and consistency, backlinks from locally relevant sites, and broader brand mentions all feed prominence. The component that practitioners consistently rank highest is reviews.
Review signals split into volume, rating, recency, and content. Volume builds prominence over time, but each additional review past a category-specific threshold adds less than the one before it. Rating contributes; profiles with average ratings below 4.0 in competitive categories struggle to rank well. Recency has become more important through 2024 into 2026: a business with 80 fresh reviews from the past 30-60 days often outranks a competitor with 500 reviews collected over years with nothing recent. Review content (the words customers use) feeds keyword-level relevance back into the prominence pillar through what the reviews describe.
Citation signals work through NAP (Name, Address, Phone) consistency across major directories. Profiles with consistent NAP across 50+ directories (Yelp, Bing Places, Apple Maps, industry-specific directories) carry stronger prominence than profiles with drift or gaps. The work is largely table-stakes in 2026: every competitor in most markets has citations in good order, so consistency keeps a profile competitive rather than producing an advantage on its own.
Link signals contribute through locally relevant sources. A link from the local Chamber of Commerce, a neighborhood blog, or a sponsored community event carries more weight for local rankings than a generic high-authority national link. A restaurant covered in the city paper’s “best new openings” piece gets a stronger local prominence lift than the same restaurant linked from a national food blog. Domain authority correlation with local pack ranking is weak compared to the relevance and proximity of the linking source.
Brand mentions across the web (especially on locally rooted sites and in news coverage) feed prominence even without backlinks. Co-occurrence of the business name with location terms produces an entity signal Google reads as prominence reinforcement. A personal injury firm mentioned by name in local court reporting and bar association directories accumulates entity signals that competitors without that footprint can’t replicate quickly. The pattern is harder to game than backlinks because it requires operational presence in the local market that can be verified externally.
Behavioral signals do what the profile alone can’t:
Engagement signals from how users interact with a profile feed back into rankings as real-world validation. Click-through rate from the SERP to the profile, click-to-call rate, direction-request rate, photo views, and post engagement all contribute to a profile’s quality score within its candidate set.
The mechanic is comparative rather than absolute. A profile that converts at higher rates than its candidate-set peers gets boosted within the local pack. A profile with mediocre engagement among strong competitors ranks lower than the absolute numbers would suggest. The relevant benchmark is the candidate set Google considers for that query in that location.
Photos are the most under-invested input. Profiles with rich photo libraries see meaningful click-rate lift compared to profiles with stock images or sparse galleries. The click-rate lift then feeds engagement signals that feed ranking. Google reads the engagement, not the visual quality of the photos themselves.
The behavioral pattern works in both directions. A profile that gets low click-through despite ranking well drifts down over time because the algorithm reads the low engagement as candidate-set mismatch. A profile that ranks at position three but earns disproportionate engagement gets pulled up over weeks as the signal accumulates.
| Signal | What it measures | What moves it | Categories where it dominates |
|---|---|---|---|
| Click-through rate | Tap rate from local pack to profile | Strong primary photo, complete profile, clear category | Restaurants, retail, medspa (visual decision) |
| Click-to-call rate | Direct call action from profile | Visible phone number, clear category match, urgency in query | HVAC, plumbing, legal (urgent service) |
| Direction requests | Map navigation initiations | Storefront-style business with physical destination | Retail, restaurants, dental (in-person visit) |
| Photo views | Tap-through on profile photos | Rich photo library with recent uploads | Restaurants, medspa, salons (visual proof) |
| Post engagement | Clicks on GBP posts | Active posting cadence with image-led content | Restaurants, retail (recurring promotions) |
A complete profile competes; an incomplete profile filters:
Profile completeness shapes whether a business is in the candidate set Google considers for a query in the first place. Profiles with missing fields, empty categories, no photos, or thin service lists get filtered out of competitive candidate sets even when the underlying business operations are strong.
The completeness check covers every field GBP exposes: primary category, secondary categories, full address, verified phone number, hours of operation, services list with descriptions, attributes appropriate to the category, photo gallery, business description, and ongoing post activity. Each empty field is a signal Google reads as an incomplete profile.
The mechanic isn’t a hard cutoff. An incomplete profile can still rank above a more-complete one if other signals (proximity, reviews, links) compensate. The pattern is that completeness compounds with other signals rather than acting independently. A complete profile makes the other signals work harder; an incomplete profile dilutes whatever signals exist.
Setup order matters: completeness gets handled before optimization begins. A half-filled profile doesn’t get optimized into ranking; the missing fields have to be populated first. Once the profile is complete, the optimization work on reviews, links, and behavioral signals has somewhere to land.
Spam filters change which profiles get to compete:
Google runs spam filters on the local pack that remove profiles violating policy before ranking even applies. Profiles with keyword-stuffed names, fake addresses, duplicate listings, suspicious review patterns, or policy violations get filtered out of the candidate set regardless of their other signals.
The filters are category-specific in their aggression. Locksmiths, addiction recovery centers, garage door repair, and other categories with historical spam problems get stricter filtering than less-targeted categories. Businesses in these categories face a higher bar for staying visible because the filter has to err on the side of removing borderline profiles to protect the index.
For legitimate businesses in heavily-filtered categories, operational hygiene matters more than for businesses in clean categories. NAP consistency, name accuracy, photo authenticity, review patterns that don’t trigger spam detection, address verification: these all have to hold up to filtering that wouldn’t catch a clean-category business with similar gaps.
Filter recovery, when a legitimate business gets caught by spam detection, runs through the GBP support process rather than through ranking optimization. A business filtered out of the candidate set can’t optimize its way back in; the filter has to be cleared first. The recovery work is administrative (verification, documentation, support tickets) rather than SEO.
Multi-location chains compete with themselves before the SERP:
Multi-location businesses face a ranking calculation single-location businesses don’t have to consider. Each location has its own GBP, its own review profile, its own service area, and its own behavioral signals that combine differently with the surrounding market than any other location’s signals do. The locations compete with each other for visibility in overlapping markets before they compete with external competitors.
The competition runs through proximity primarily. A franchise with three locations in a city ranks each location strongest near its address and weaker as distance grows. In areas between two locations, neither location ranks as strongly as it would if it were the only nearby option. The brand is more visible overall, but each individual location loses ground to its sibling locations in overlap zones. An HVAC franchise with locations in Park Slope, Williamsburg, and Greenpoint sees each location dominate its own neighborhood and weaken in the streets between them; a dental practice with two offices on opposite sides of a city avoids this overlap entirely.
The pattern shifts strategy for multi-location operations. Each location is treated as its own ranking project rather than a copy of headquarters. Local managers handle reviews and posts for their location; central marketing handles consistency of branding, NAP, and structured information. The mistake to avoid is treating all locations as a single broadcast unit.
For service-area businesses with multiple locations covering large territories, the service-area boundary setup matters more than for storefronts. Two locations with overlapping service areas compete for the same queries; locations with cleanly drawn non-overlapping areas don’t.
The audit pattern for multi-location operations is comparative: across all locations, which rank well, which underperform, and what explains the difference? Locations with similar offerings ranking differently usually point to specific signal gaps (review velocity drift, profile drift, lost citations) rather than to category-level issues.
What’s new in 2026: AI Overviews and what stayed the same:
The biggest change in 2026 local search has been the integration of AI Overviews above the local pack for an increasing share of queries. AI Overviews synthesize information from business descriptions, reviews, schema, and website content into a generated summary that appears above the traditional 3-pack. The Overview pulls representative businesses, ratings, and review snippets from a candidate pool that overlaps heavily with the 3-pack candidates but uses similar underlying signals.
The work that lifts a business into the 3-pack also tends to lift it into AI Overview citation. Well-cited businesses with strong review profiles dominate both surfaces. The two layers are converging rather than requiring separate optimization stacks.
What hasn’t changed is the underlying ranking signal architecture. Google’s documentation still describes the relevance/distance/prominence triad. GBP completeness still carries the heaviest single category of influence. Reviews still feed both prominence and relevance through volume, recency, and content. Citations still serve as table-stakes consistency requirements rather than as ranking levers themselves.
What has shifted is the relative weight of recency signals. Through 2024 into 2026, recency in reviews and recency in profile activity (posts, photos, updates) carry more weight than they did in earlier years. A profile with steady recent activity outranks a profile with stronger absolute numbers but no recent updates. The freshness mechanic that always applied to posts now extends through more of the profile.
Local SEO fundamentals (verified profile, complete information, clean citations, active review program, locally relevant links) still produce ranking outcomes. AI Overviews didn’t replace the work. They added a surface that draws from it.
The three pillars compound; missing any of them caps the rest:
The local 3-pack ranking system rewards businesses that combine relevance (matching the query through profile completeness and category accuracy), proximity (being physically near the searcher or the queried location), and prominence (carrying reputation signals that exist outside the profile). The three factors compound; missing any of the three caps the contribution of the other two.
What makes local ranking work isn’t optimizing one factor to its maximum. The threshold matters more than the maximum. It’s making sure all three reach the threshold that lets them combine. A perfectly optimized profile in a category the business doesn’t legitimately occupy fails relevance. A business in the right location with the right category but no review velocity fails prominence. A business with strong reviews and the right category at a remote address fails proximity for most of its potential queries.
Relevance compounds, distance softens, prominence builds; the leverage is the intersection. A business that gains in one pillar lifts the others through behavioral and entity signals that feed back through the system. The competitor displacement test (identify the current top three and compare three-pillar standing) reveals which lever to pull first, which is usually the pillar the business is furthest behind in for the specific market and queries it cares about.