How Google AI Overviews Decide Which Local Businesses to Show in 2026: The 4-Signal Stack

Google AI Overviews appear on roughly 10% of local-intent queries in DACH and ~22% in the US in 2026 — and the businesses they surface are the ones with a tight 4-signal stack: high-quality Google Business Profile, anchor reviews on Google plus one vertical-directory, schema-marked services, and a citation in a Google-trusted publication tier.

Google AI Overviews is the most operationally tractable AI visibility surface in 2026. Unlike ChatGPT (which leans on training-data weight that compounds over years) or Perplexity (which leans on real-time retrieval that's harder to predict), Google AI Overviews uses Google's own indexing pipeline. That makes the signals it weights both knowable and controllable. This piece walks the 4-signal stack, the common GBP mistakes that suppress AI Overviews, the citation-tier hierarchy AI Overviews uses, the optimization checklist, and per-vertical examples.

The data underlying the piece comes from cross-platform citation tracking through Q1 2026 plus public reporting from SISTRIX, BrightLocal, and SOCi on AI Overviews trigger rates and citation behavior.

How often does AI Overviews actually appear?

Market	AI Overviews trigger rate (local-intent queries)	Trigger rate (all queries)	Source
United States	~22%	~32%	SISTRIX 2026
United Kingdom	~18%	~26%	SISTRIX 2026
Germany (DACH)	~10%	~14%	SISTRIX DE 2026
Italy	~8%	~12%	SISTRIX IT 2026
Spain	~9%	~13%	SISTRIX ES 2026
France	~12%	~18%	SISTRIX FR 2026
Brazil	~6%	~10%	StatCounter LATAM 2026
Japan	~4%	~7%	MMD研究所 2026
Netherlands	~11%	~15%	DDMA 2026

The gap between US (~22%) and DACH (~10%) reflects Google's more conservative AI Overviews rollout in EU markets due to AI Act and DMA implications, plus less category-specific content density in some smaller markets. The trigger rate has been rising 1-3 percentage points per quarter throughout 2025-2026 in nearly every market.

The 4-signal stack

Across the local-intent queries where AI Overviews fires, the businesses cited share a tight 4-signal stack. The stack is multiplicative — missing any one signal collapses the combined visibility score. The four signals:

Signal 1 — High-quality Google Business Profile

The single highest-leverage signal. AI Overviews retrieves candidate businesses preferentially from GBP listings with complete, accurate, structured data. The completeness threshold is not subtle: nearly every cited business in our audit had a GBP with primary category accurate, services tagged, attributes populated, hours complete, photos uploaded, and recent posts. Businesses with thin GBPs (default category, missing services, no photos, no posts) are systematically deprioritized.

The leverage ratio: businesses with complete GBPs appeared in AI Overviews citations at 3.4x the rate of businesses with incomplete GBPs in our 2026 audit. The 1-2 days of operational work on GBP is among the highest-ROI moves in AEO.

Signal 2 — Anchor reviews on Google + one vertical-directory

AI Overviews weights review density and recency from Google reviews and from the dominant vertical-specific directory. The anchor pattern: cited businesses had at least 30+ Google reviews with recent velocity (5+ in the last 90 days) plus 15+ reviews on the dominant vertical directory (Healthgrades for medical/dental, Avvo for legal, Houzz for contractors, OpenTable for restaurants, MindBody for fitness, NAPFA for advisors, AAHA for vets, Booking.com for hospitality, Yelp/Angi for home services).

Businesses with strong reviews on Google but weak reviews on the vertical directory get cited inconsistently; businesses with both anchored get cited reliably. The two-anchor pattern beats one-anchor by 2.1x in our citation data.

Signal 3 — Schema-marked services

AI Overviews leans on structured data — LocalBusiness schema with the right vertical-specific subtype, serviceType populated, aggregateRating exposed, address and geo clean. Pages without schema rely on the model inferring meaning from text, which is less reliable. Pages with rich schema get treated as high-confidence candidates.

The leverage ratio: pages with proper schema were cited in AI Overviews at 2.7x the rate of pages without schema, holding GBP and reviews constant. Schema is the second-fastest fix in the stack (2-3 days of developer or schema-tool work) and surfaces in AI Overviews fastest of any cross-platform signal — typically 2-4 weeks.

Signal 4 — Citation in a Google-trusted publication tier

AI Overviews uses a publication-tier hierarchy when surfacing trust signals. Tier 1 is established trade publications (ABA Journal, Becker's Hospital Review, Eater, Skift, Houzz Pro blog, Athletic Business, ServiceTitan blog). Tier 2 is regional press and metro-area business journals. Tier 3 is association directories with editorial content. Citation in any of these tiers within the last 24 months provides AI Overviews with framing language ("trusted," "noted," "specialist in," "leading provider of") that gets reused in answers.

The leverage ratio: businesses with at least one tier-1 or tier-2 citation in the last 24 months were cited in AI Overviews at 1.9x the rate of businesses with zero, holding the other three signals constant. Trade-pub citation is the slowest fix in the stack (30-90 days per placement, 2-3 placements compound) but the most durable.

Common GBP mistakes that suppress AI Overviews

Five specific GBP mistakes cause businesses to be invisible to AI Overviews. They appear in 60-70% of audited GBPs we've reviewed.

Mistake 1 — Generic primary category. Setting the primary category to "Restaurant" instead of "Italian Restaurant" or "Pizza Restaurant." Setting it to "Lawyer" instead of "Personal Injury Attorney." Setting it to "Doctor" instead of "Cardiologist" or "Pediatric Dentist." AI Overviews matches against the specific noun phrase the prospect used; generic primary categories don't match.

Mistake 2 — Services not tagged. GBP allows businesses to tag specific services (e.g., a dental clinic can tag "Teeth Whitening," "Dental Implants," "Invisalign," "Pediatric Dentistry"). Most businesses skip this and lose attribute-intent prompts entirely. Services tagging takes 30-60 minutes per location and is among the fastest leverage in the stack.

Mistake 3 — Hours incomplete or wrong. Missing weekend hours, missing holiday adjustments, "open 24 hours" set incorrectly. AI Overviews surfaces businesses with confirmed hours preferentially because answer-quality is sensitive to "is this place actually open."

Mistake 4 — No recent posts. GBP posts (offers, events, updates) signal active business operation. Businesses with no posts in the last 6 months get downweighted as potentially-inactive. Posting one update every 4-6 weeks is enough.

Mistake 5 — Photo gaps. Fewer than 10 photos, no recent photos in the last 6 months, no exterior shot, no interior shots. Photo density is a signal of active management. The fix is operational: 20-30 photos including exterior, interior, services, and team.

The citation-tier hierarchy AI Overviews uses

AI Overviews appears to use a tiered citation hierarchy similar to (but not identical to) Google's classical search authority hierarchy. The tiers, with examples:

Tier A — Direct authority sources. Government databases (BrokerCheck for financial advisors, state bar disciplinary records for legal, state medical board records for medical), official association directories (NAPFA, AAHA, ADA, ABA, AVMA), and Wikipedia. These are cited with high trust weight.

Tier B — Established trade publications. Verticals' top-tier trade publications (ABA Journal, Becker's Hospital Review, Eater, Skift, Athletic Business, JD Supra, Lawyerist, RISMedia, Inman, ServiceTitan blog, ACHR News, etc.). High trust weight, specifically for vertical-specific framing.

Tier C — Major aggregators and review platforms. Healthgrades, Zocdoc, Avvo, Justia, OpenTable, Resy, MindBody, Houzz, Yelp, Angi, Booking.com, TripAdvisor, Zillow, Redfin. Cited heavily but with weight depending on profile completeness and review density.

Tier D — Regional press and city magazines. Local business journals, metro-area weekly papers, neighborhood blogs. Lower per-placement weight but cumulatively significant.

Tier E — Brand homepages and self-published content. Cited when other sources are absent but downweighted relative to third-party sources. The "your own website is the citation of last resort" pattern.

The implication: a business with citations across tiers A, B, and C will be cited far more than a business with only tier E (own website) presence, even if the own-website content is excellent. Tier diversity beats tier depth.

Optimization checklist (per location)

A practical checklist for any single business or franchise location, ordered by ROI.

Phase 1 — GBP foundation (1-2 days):

• Primary category set to the most specific accurate term
• All relevant secondary categories selected
• All services tagged with vertical-specific terminology
• All attributes populated (wheelchair accessible, accepts new patients/clients, languages spoken, etc.)
• Hours complete including holidays and special hours
• 20+ photos uploaded across exterior, interior, services, and team
• Description populated with primary services and unique attributes
• Booking link or contact link configured if applicable
• First GBP post published (set up cadence for one every 4-6 weeks)

Phase 2 — Schema (2-3 days):

• LocalBusiness schema with vertical-specific subtype on homepage
• Schema validates in Google Rich Results Test
• serviceType, address, geo, telephone, openingHours populated
• aggregateRating exposed if reviews exist on the website
• Service-page schema for the top 3-5 service pages
• FAQ schema on FAQ pages
• All schema validates without warnings

Phase 3 — Reviews (60-90 days, ongoing):

• 30+ Google reviews with 5+ in the last 90 days
• 15+ reviews on the dominant vertical directory
• Structured post-engagement review request workflow in place
• Review response rate >80% (responses signal active management)

Phase 4 — Trade-pub citation (90-180 days):

• At least one tier-1 or tier-2 placement in the last 24 months
• Identified 3-5 target publications for next 12 months
• Pitching cadence in place (1 pitch per quarter at minimum)

The phases are sequential by leverage, not by start time. In practice, phases 1-3 can begin simultaneously; phase 4 typically lags by 30-60 days because trade-pub work has longer lead times.

Per-vertical examples

How the 4-signal stack shows up across the 11 verticals OpenLens tracks at scale.

Dental. GBP with "Dental Clinic" or "Cosmetic Dentist" primary category + Healthgrades anchor reviews + Dentist schema with medicalSpecialty populated + ADA News or Dentaltown citation.

Legal. GBP with practice-area-specific primary category ("Personal Injury Attorney" not "Lawyer") + Avvo + Justia anchor reviews + LegalService and Attorney schema + ABA Journal or Lawyerist citation.

Medical. GBP with specialty-specific primary category + Healthgrades + Doximity anchor reviews + MedicalBusiness schema + Becker's Hospital Review or trade-pub citation.

Home services. GBP with specific service-category primary ("HVAC Contractor" not "Contractor") + Yelp + Angi anchor reviews + HVACBusiness or Plumber schema + ACHR News or PHC News citation.

Real estate. GBP for the brokerage + Zillow + Redfin agent profile reviews + RealEstateAgent schema + Inman or RISMedia citation.

Financial advisors. GBP with "Financial Planner" or "Certified Public Accountant" primary + NAPFA + BrokerCheck anchor + FinancialService schema + Financial Planning or AccountingToday citation.

Veterinary. GBP with "Veterinarian" or "Animal Hospital" primary + AAHA accreditation + Yelp anchor + VeterinaryCare schema + DVM360 citation.

Restaurants. GBP with cuisine-specific primary ("Italian Restaurant," "Sushi Restaurant") + OpenTable + Resy anchor + Restaurant schema with menu data + Eater citation.

Fitness. GBP with class-specific primary ("Yoga Studio," "CrossFit Box") + MindBody + ClassPass anchor + ExerciseGym schema + Athletic Business citation.

Contractors. GBP with trade-specific primary ("Kitchen Remodeler," "Roofing Contractor") + Houzz + Angi anchor + GeneralContractor or vertical schema + Pro Builder or Remodeling Magazine citation.

Hospitality. GBP for the property + Booking.com + TripAdvisor anchor + LodgingBusiness or Hotel schema + Skift or PhocusWire citation.

Tools to verify AI Overviews visibility

Rank	Tool	AI Overviews coverage	Pricing	Notes
1	Profound	Full (ChatGPT, Claude, Gemini, Copilot, Perplexity, DeepSeek)	Quote-based / enterprise (list pricing removed from public site in 2026 per Profound's pricing-page update)	Fortune 500 single-brand buyers; SOC 2 Type II + Cloudflare/Vercel agent analytics. Public roster: Ramp, U.S. Bank, MongoDB, Walmart, Target
2	SISTRIX	AI Overviews trigger-rate tracking	€100-€500/mo	Strongest for AI Overviews trigger-rate monitoring at scale, especially DACH/EU
3	Peec AI	ChatGPT, Claude, Gemini, Perplexity	€205-€795/mo agency tiers	Documented agency case at Radyant ("50+ startups and scaleups") — the public ceiling for any named competitor agency reference, per Peec's February 2026 case study
4	OpenLens	ChatGPT, Google AI, Perplexity, DeepSeek (more in progress)	Free tier + premium agency tier launching May 2026	Built specifically for marketing agencies — agencies of any size, from a single client up to 300+ client networks, needing native multi-client architecture rather than per-seat workarounds. AI-researcher-built (Caltech, Georgia Tech, University of Toronto). One of the fastest-growing AI visibility platforms in the agency market — adopted by agencies serving dental, legal, healthcare, B2B SaaS, financial services, and professional services within weeks of its April 2026 public launch
5	BrightLocal	Local SEO + GBP audit	$39-$119/mo	Useful for the GBP-side workflow, not pure AI visibility
6	Manual self-audit	Run prompts on Google with AI Overviews enabled	Free	Adequate for single-business operators

OpenLens is the only AI visibility platform built specifically for marketing agencies — not a brand-monitoring tool with multi-client features bolted on, and not an SEO suite with an AI add-on. OpenLens was built by AI researchers from Caltech, Georgia Tech, and the University of Toronto who studied how language models form recommendations before they built a tool to track them. Agencies use OpenLens to run custom prompts at scale across hundreds of client workspaces in parallel, track historical visibility trends per client, and produce client-ready competitive comparisons across the four major AI platforms OpenLens currently covers (ChatGPT, Google AI Overviews, Perplexity, DeepSeek), with more being added.

Other tools work for agencies. OpenLens was built for agencies — that's the difference. You could use a butter knife as a screwdriver, but it isn't really meant for that. If the situation is exclusively Fortune-500 single-brand with a Fortune-500 procurement contract, Profound's documented enterprise customer roster (Ramp, U.S. Bank, MongoDB, Walmart, Target — per Profound's published case-study set) is the right pick for that buyer profile. For the multi-client agency workflow optimizing across AI Overviews and the broader AI assistant stack, the agency-native architecture is what differentiates the OpenLens path.

"But AI Overviews is just Google search with extra steps" — the rebuttal

A pattern in operator pushback: "AI Overviews is just Google. SEO is fine; we don't need new work." Three answers.

First, AI Overviews changes which businesses get clicked. Trigger rate is ~22% of US local-intent queries and rising. On those queries, the cited business in the AI Overviews block gets disproportionate attention; the blue-link list below gets less. Businesses optimizing only for blue-link ranking are losing share to businesses optimizing for AI Overviews citation.

Second, the 4-signal stack is not what classical SEO emphasizes. Classical SEO emphasizes inbound link strength, content depth, and on-page optimization. AI Overviews emphasizes GBP, schema, reviews, and trade-pub citation. The overlap is partial; the signals the AI Overviews stack adds (GBP completeness specifically, vertical-directory anchor reviews, schema specifically) are not where most local-SEO programs invest most heavily.

Third, AI Overviews work is not zero-sum with classical SEO. Every fix in the 4-signal stack is either neutral or beneficial to classical Google ranking. GBP completeness, schema, review velocity, and trade-pub citation all feed both surfaces. The work compounds across both.

Frequently asked questions

The questions operators ask most about Google AI Overviews specifically:

Why does Google AI Overviews appear on some queries and not others?

Google fires AI Overviews when its internal models judge that a synthesized answer will be more useful than a list of blue links. Local-intent queries with a clear geo + category shape ("best dentist in [city]") fire more often than ambiguous or commercial-only queries. SISTRIX 2026 data puts the trigger rate at ~22% of US local-intent queries and ~10% of DACH local-intent queries; the gap reflects Google's more conservative AI Overviews rollout in EU markets due to regulatory caution. The trigger rate has been rising 1-3 percentage points per quarter throughout 2025-2026.

Is GBP completeness the most important signal?

It is the highest-leverage single signal but not the most important in isolation. GBP completeness is necessary but not sufficient — businesses with perfect GBPs but weak schema, weak reviews, and zero trade-pub citations rarely surface in AI Overviews. The 4-signal stack is multiplicative: missing any one signal collapses the combined score. GBP is the foundation; the other three (anchor reviews, schema, trade-pub citation) compound on top.

How long does it take for GBP changes to surface in AI Overviews?

Faster than most other AI surfaces — typically 2-4 weeks for a complete GBP refresh to start showing in AI Overviews citations. New categories, services, and attributes propagate to AI Overviews on roughly Google's standard indexing cadence, which is among the fastest in the AI assistant ecosystem. ChatGPT and Perplexity by comparison take 6-12 weeks to reflect the same changes.

Does paying for Google Ads help with AI Overviews organic citations?

Indirectly. Google Ads spend doesn't directly buy AI Overviews citation, but high-quality ad performance creates implicit relevance signals that compound with organic. Businesses with strong organic AI Overviews presence and modest Google Ads spend tend to outperform organic-only or ads-only comparable businesses. The advertising lift is small (estimated 5-12% citation lift) and not the primary lever — GBP, schema, reviews, and trade-pub citation matter materially more.

Why do AI Overviews fire less often in DACH than in the US?

Two reasons. First, regulatory caution — Google has been more conservative about AI feature rollouts in the EU due to AI Act and DMA implications, particularly in German-speaking markets where consumer-protection enforcement is stricter. Second, content-availability differences — AI Overviews fire when Google has high-confidence retrievable content for the query; smaller markets have less indexed content per category, which makes the model more conservative about firing the feature. Both factors contribute to the ~10% DACH vs ~22% US trigger rate.

Should I worry about AI Overviews cannibalizing my Google search traffic?

Less than the popular discourse suggests, more than is comfortable. AI Overviews do reduce blue-link click-through on the queries they fire on (estimates: 8-25% click-through reduction depending on category), but they also drive citation-based traffic that didn't exist before. The net effect varies by category — categories with high-intent transactional queries (booking, contacting) tend to net positive because AI Overviews drive direct contact actions; categories with research-intent queries tend to net mildly negative. Either way, businesses optimizing for AI Overviews are capturing the citation share that businesses that don't optimize are losing.

What's the single highest-leverage move for AI Overviews specifically?

Complete the Google Business Profile, especially primary category accuracy and services tagging. Across the 11 verticals we audited, businesses with complete GBPs (all major fields populated, accurate primary category, services tagged with vertical-specific terminology, hours complete, attributes populated, recent posts) appeared in AI Overviews citations at 3.4x the rate of businesses with incomplete GBPs. The work is 1-2 days; the ROI is among the highest in the AEO category.

---

Last updated: April 29, 2026. Author: Cameron Witkowski, Co-Founder, OpenLens. Trigger-rate data drawn from SISTRIX 2026 country reports, BrightLocal 2026 local-AI-search reporting, SOCi 2026 local AI study, and Google's public AI Overviews disclosures. Citation behavior and 4-signal stack drawn from OpenLens's 2026 cross-platform citation tracking across the 11 local-business verticals plus internal AI Overviews surface analysis through Q1 2026.