AI Visibility
for local business
How AI search actually works — and what you can do about it.
Perplexity Cites Your Homepage, Not Your Listicle: Home Services Data
The standard AEO playbook says create listicle content for Perplexity. A 25,337-citation study says that's wrong for HVAC companies, plumbers, and electricians.
ChatGPT Brand Citations Are Recovering — but Only for Paying Users
GPT-5.6 Sol reversed ChatGPT's brand-site citation collapse — but Sol only runs for paid users. Free-tier customers still see GPT-5.5 Instant behavior, and most SMB buyers are on the free tier.
Why Your AI Visibility Fix Works on One Platform and Fails on Another
Yext's 17.2M citation study reveals that ChatGPT cites from directories, Gemini from brand sites, and Perplexity from niche expert sources. Each platform needs a different fix.
How Cloudflare's New Default Could Silently Block AI From Seeing You
On September 15, new websites on Cloudflare will have AI training and agent bots blocked by default. Here's what that means -- and what to check before it affects your visibility.
GPT-5.6 Sol Is Seeking Out Your Website Again
GPT-5.5 dropped ChatGPT's brand-site citation rate to near zero. A July 2026 investigation tracks a Writesonic finding that GPT-5.6 Sol may have reversed that completely — for paid users.
What Sol's '97% Citation Accuracy' Actually Measures (And What It Doesn't)
AEO practitioners are citing Sol's developer benchmark as evidence of a web citation advantage. The number is real. The application is wrong.
GPT-5.6 Sol vs. Terra: What the Citation Numbers Actually Measure
Practitioner content is citing Sol/Terra citation rate differences with confidence. Our July 17 investigation found those numbers measure the wrong thing entirely.
GPT-5.6 Sol, Terra, and Luna: What's Actually Known About Citation Rates
GPT-5.6 ships three models -- Sol, Terra, Luna -- each reaching different users. Here's what's confirmed about their citation behavior and what isn't yet.
Why the ChatGPT Your Customers Use Cites Differently Than Your Test
GPT-5.6 serves free, Plus, and Pro users three distinct models with an estimated 8x citation gap between tiers. What you see in your test may not match what your customers see.
Why Local Pack Optimization Doesn't Transfer to AI Search
The same signals drive both -- but at completely different weights. New Whitespark data makes the divergence explicit for the first time.
SiriKit Is Deprecated. Your Booking Software Has a New AI Access Problem.
iOS 27 public beta confirmed SiriKit's deprecation on July 13. Booking software that hasn't migrated to App Intents now has a 2-3 year clock before losing Siri AI access entirely.
Local Businesses Start at an 11% AI Citation Rate. Here's the Path to 44%.
A 2026 study of 100,000+ AI prompts found local businesses have an 11% baseline citation rate. July 2026 research identifies what moves the number -- and the review platform gap most miss.
The AI Citation Moat Vendors Are Selling Doesn't Exist
A claim spreading through AEO vendor content this week says GPT-5.6 prompt caching creates a citation moat for early-mover brands. We investigated. The mechanism doesn't exist.
Gemini Is Now the #2 AI Traffic Source. Here's What Changed in January 2026.
Gemini grew from 4.3% to 13.2% of AI referral traffic in one quarter. A January model transition rewrote 42% of citation pools -- and what drives Gemini visibility has fundamentally shifted.
Why the Same Business Appears in ChatGPT for One User but Not Another
ChatGPT's extended thinking mode cites 74% different sources than its default. Combined with citation non-determinism, this explains audit results that seem to contradict what users see.
Yelp Captured 62% of Perplexity Citations. Why Review Freshness Explains It.
Across 28.5 million AI queries, Yelp captured 62% of Perplexity citations and 73% of Google AI Mode's. The mechanism -- and what it means for your review program.
ChatGPT and Perplexity Can't Read Your Google Reviews
Google Business Profile renders review content via JavaScript -- external AI crawlers can't read it. The platforms that affect ChatGPT and Perplexity visibility are Yelp and BBB.
There's No Submit Button for ChatGPT. What Actually Gets You Cited.
Scammers are charging businesses $99-$499 to submit them to ChatGPT. The US SBA warned about it in June 2026. Here's why it's impossible and what actually gets you cited.
The Two Review Thresholds That Control AI Citation Eligibility
AI engines apply two independent review filters: a quality floor and a volume floor. Clearing one without the other still leaves you invisible.
Ask Maps Doesn't Rank Contractors. It Picks One.
Our July 2026 investigation found Ask Maps returns one named contractor with reasoning -- not a list. Here's what drives the pick, and how long it takes to get there.
Which AI Platform Actually Sends Traffic in 2026
Gemini has 27% of AI users but only 12% of referral traffic. ChatGPT has 52% of users and 80% of referral traffic. The two metrics don't move together -- and the gap changes how you should prioritize platforms.
Why Expert Best-Of Lists Are the #1 AI Visibility Factor
BrightLocal's 2026 ranking survey put expert 'Best Of' list presence at #1 for AI visibility. Most businesses have zero. Here's why it matters and the two-layer fix.
Why ChatGPT Local Recommendations Run Through Foursquare, Not Yelp
60-70% of ChatGPT local business results trace back to Foursquare's Places API, not Yelp or Google. Most businesses haven't updated their Foursquare listing since it was a consumer app.
Your GBP Is Feeding Three AI Systems Now. Most Guides Describe One.
GBP Q&A seeding is dead. Three new AI systems read your profile instead: Ask Maps, Ask for Me, and AI Mode. Each has requirements that didn't exist six months ago.
GBP Q&A Is Gone. Here's What Google Replaced It With.
Google killed GBP Q&A in late 2025. Ask Maps -- a Gemini-powered AI answering layer -- replaced it. The optimization model is completely different.
The Selection Filter Google Runs Before Ask for Me Dials a Contractor
Google's 'Ask for Me' filters contractors before dialing. A GBP listing 'Plumbing Services' won't get called about water heater replacement. The exact service needs a Services tab entry.
Google's AI Calls Contractors for Homeowners — Here's Who Gets Picked
Google's 'Ask for Me' feature calls 3-5 contractors on homeowners' behalf for pricing. Invoca tracked a 300%+ call volume spike in its first month. 26% of businesses didn't answer.
44% of ChatGPT Citations Come From the First Third of the Page
A study of 3 million ChatGPT responses found 44.2% of citations come from the first 30% of page content. Where the answer sits is now a primary citation variable.
Perplexity Searches Literally. ChatGPT Rewrites Everything.
Perplexity keeps 88% of original query words. ChatGPT generates 91% unique sub-queries from the same prompt. That split means vocabulary precision fixes one platform and does nothing for the other.
What the ChatGPT Citation Collapse Actually Means for Local Service Businesses
ChatGPT's source footnotes dropped 86-94% this spring. But local recommendation queries work differently -- and the fix that follows is not what most guides recommend.
AI Citations Change Every Month -- Even When You Don't
40-60% of the sources AI platforms cite for identical queries change every month. Here's the platform-by-platform retention data -- and what it means for your audit cadence.
Reddit Dominates ChatGPT Citations. Local Services Are the Exception.
Reddit grew 59% on ChatGPT after GPT-5.5, but that data covers informational queries, not local service searches. Q4 2025 citation data shows Yelp at 512,680 local citations. Reddit doesn't rank in the top six.
75% of Home Service AI Searches Are Emergency Queries. Audits Miss Them.
When a pipe bursts, homeowners don't type 'best plumber Dallas.' They type 'burst pipe emergency plumber near me.' New 5W data shows 75% of home service AI searches are incident-framed -- a query type most audits ignore.
ChatGPT Screens Out Businesses Below 4.0 Stars Before Reading Anything
SOCi's 2026 study of 350,000 locations found ChatGPT excludes businesses below 4.0 stars before evaluating infrastructure signals. Schema and directories don't help until the threshold is cleared.
Why Perplexity Ignores Your Content Fixes (And What Works Instead)
FAQ schema, structured content, dateModified updates -- Perplexity still shows zero. The reason isn't content quality. It's that you cleared the wrong bar.
Delta Gets More AI Citations Than American Airlines
Delta holds 10.5% AI citation share for airline queries. American flies the most domestic seats. The same decoupling runs through hotels and home services.
How to Write FAQ Answers That AI Search Engines Actually Cite
Most FAQ sections are written for humans who skim them. AI search engines need three things: buyer incident language, answers of 40-80 words, and the complete answer in sentence one.
AI Citation Lock-In: Why Early Movers in Local Search Are Hardest to Displace
AI citation patterns compound once formed. The first businesses to build entity infrastructure in each local market are structurally harder to displace than latecomers.
GPT-5.5 Cut Brand Site Citations by 55%. Here's the New Playbook.
On May 23, ChatGPT's default model shifted to GPT-5.5 Instant. Brand site citations on the free tier fell 55%. We tracked the mechanism and what works instead.
Apple's Siri AI Books Service Appointments. The Same Setup Covers Google Too.
Apple's Siri AI uses the same Housecall Pro and ServiceTitan endpoints as Google AI Mode. One booking integration now gates access to both platforms simultaneously.
What Google's New GSC AI Performance Report Actually Tells You
Google added AI impressions data to Search Console on June 3. Here's what the report shows, the opt-out trap that silently zeros your numbers, and how to use it as a measurement baseline.
Shopify Is Already AI Agent-Ready. Service Businesses Need a Different Check.
Chrome desktop auto-browse has been live since May 19. The infrastructure AI agents need to transact is completely different for Shopify stores vs. service businesses.
Why 87% of Independent HVAC and Plumbing Contractors Have Zero AI Citations
87% of independent HVAC and plumbing contractors have zero measurable AI citation share. Our June 2026 research shows reviews aren't the fix — entity infrastructure is.
When AI Agents Book Appointments: Three Infrastructure Checks That Matter
Google's Chrome Android agentic booking layer ships late June 2026. Citation presence isn't enough — three infrastructure checks now determine whether the agent completes the transaction or bounces.
The Two-Phase AI Citation Problem New Businesses Don't Know About
Most AI citation advice assumes you're already in the citation pool. Newer businesses aren't -- and running Phase 2 tactics before Phase 1 is solved produces zero movement.
Why Perplexity Shows Your Fixes First (And Claude Takes Six Weeks)
AI platforms have different retrieval architectures, and those architectures determine how quickly your fixes propagate. A platform-by-platform breakdown.
ChatGPT Doesn't Pull From Google. It Pulls From Bing.
87% of ChatGPT's cited sources overlap with Bing's top-10 results -- not Google's. Here's what our June 2026 research found and what to do about it.
Why Generic Schema Won't Help Your AI Citations
Two published studies on schema and AI citations appear to contradict each other. Here's what the data actually says — and what it means for SMBs starting from zero.
When Perplexity Doesn't Know Your Business Exists
Two audit findings, two failure modes. Both look like you're invisible on Perplexity -- but the causes are completely different and need completely different fixes.
Microsoft's New AI Citation Report: Useful, But Missing 80% of the Picture
Bing Webmaster Tools now shows which of your pages Copilot cites as grounding sources. It's the first free first-party AI citation data. Here's what it tells you -- and where it stops.
Your AEO Score Might Be Measuring the Wrong Thing
Three measurement gaps in standard AEO audits that produce misleading scores — what our 2026 research found, and what it means for your fix plan.
Why Claude Prefers Your Competitor's 2022 Article Over Your New Page
Claude's citation behavior is measurably different from ChatGPT and Perplexity: it skews toward older, premium sources and penalizes poorly-structured paragraphs. Here's the data.
Google SEO No Longer Predicts AI Citations
A May 2026 study of 680 million citations found the overlap between Google top rankings and AI-cited pages has fallen from 70% to under 20%. Here's the data -- and what's driving the split.
Your ChatGPT Fix Doesn't Help Perplexity: The 11% Problem
Only 11% of domains cited by ChatGPT also appear in Perplexity's citation pool for the same queries. Here's what our research shows about why -- and how to close each gap separately.
The AI Search Category Trap: When Zero Means the Wrong Race
A 0.0 Category Authority score often isn't a failure -- it's a measurement against the wrong competitors. Three audits confirmed the pattern.
ChatGPT Started Linking to Brands in Its Answers. What Gets You the Link.
On May 7, ChatGPT began inserting organic links to brand homepages in 29% of business mentions. The prerequisite is getting mentioned at all.
ChatGPT Now Links to Businesses in Answers. Who Gets the Link?
On May 7, ChatGPT started inserting clickable links to business sites inside its answers. Referral traffic jumped 157%. Here's who benefits and why.
Why Relevant Businesses Don't Appear in AI Answers
Academic research from 2026 found 43% of topically relevant pages get no AI citations -- not because they're irrelevant, but because of three distinct failure modes.
The AI Citation Bias That No Schema Fix Can Change
We audited AI platforms against each other and found a recommendation bias baked into model training. Here's what it means for your business.
AI Is Crawling Your Site. That's Not Why You Get Recommended.
AI bot traffic to small business websites grew 4x from January to April 2026. Most of those businesses still aren't getting cited. Crawling and citation are separate pipeline stages.
Google AI Mode Is Not the AI Search You've Been Optimizing For
Google AI Mode uses different signals than ChatGPT, Perplexity, or Gemini. Our May 2026 research found 13.7% URL overlap between AI Mode and AI Overviews. Your AI score may miss it entirely.
43% of Relevant Pages Get Zero AI Citations. Generic Fixes Miss the Point.
A March 2026 paper put a number on the AI citation problem: 43% of relevant pages get zero citations. Targeted diagnosis produced 40% better results than generic content rewriting.
The First Controlled Schema Study: What It Means for AI Visibility
Ahrefs tracked 1,885 pages that added schema markup. AI citation rates barely moved. Here's what the data actually says — and what drives citations instead.
68% of AI Citations Go to 15 Websites. The Other 32% Is Where You Compete.
The AI citation landscape is more concentrated than you think. Understanding the concentration tells you exactly where small businesses can actually win.
Two Ways Your AI Visibility Fix Plan Fails Before It Starts
Not every AI visibility fix fails because the recommendations are wrong. We found two structural problems that make fixes useless -- or actively harmful -- before they're implemented.
Which Directories Triple Your AI Citation Rate
An SE Ranking primary study found domains on G2, Trustpilot, and Yelp cite at 3x higher rates on ChatGPT. But Perplexity reads those same directories differently.
AI Gets Business Facts Wrong 5-20% of the Time. SMBs Are More Exposed.
Enterprise AI accuracy research puts the factual error rate at 5-20%. Small businesses with thin entity signals are almost certainly doing worse -- and rarely know it.
We Checked the Source on the Most-Cited AEO Statistics
The '85%' AI freshness figure everyone quotes is 79%. The '3.7x original research' claim has no primary study. We traced them back so you don't have to.
Why Content Freshness Matters More on Perplexity Than ChatGPT
For Perplexity, content age is a primary citation signal. For ChatGPT, it barely matters. Getting this backwards means fixing the wrong platform.
Ranking #1 on Google Used to Predict AI Citations. Now It Barely Does.
Seven months ago, 76% of pages cited in Google AI results also ranked in the top 10. Now it's 38%. The mechanism behind this shift changes how AI visibility actually works.
Why Your AI Visibility Score Doesn't Tell You What to Fix
The AEO industry has built great tools for tracking AI visibility trends. What most don't do: diagnose why you're invisible or what specifically to change.
What Perplexity Reads When It Recommends a Business
We analyzed 102 actual Perplexity source URLs from a live audit. Here's what Perplexity reads for category queries -- and it's not your website.
AI Recognizes Your Brand. That's Not the Same as Recommending You.
Brand queries and category queries are two separate AI visibility problems. Most businesses have only solved the first one -- often without realizing the second one exists.
ChatGPT Trusts Directories. Gemini Trusts Your Site.
ChatGPT pulls half its citations from third-party directories. Gemini pulls half from brand-owned sites. One fix won't work across all three -- here's what each platform actually trusts.
YouTube Overtook Reddit for AI Citations. What Local Businesses Should Do.
Reddit lost 86% of its AI citations after the 2025 lawsuit. YouTube stepped in -- and the data about which videos get cited will change how you think about the platform.
The Content Fix That Peer-Reviewed Research Actually Backs
Most AEO advice runs on vendor figures no one can trace. Princeton's peer-reviewed research found one content intervention that consistently moves the needle for AI visibility.
How to Optimize Your About Page for AI Citations
Most businesses write their About page for humans. AI models read it as an entity verification document. Here's how to make it work for both.
When AI Gives Your Business Three Different Identities
Entity fragmentation is when AI models produce contradictory descriptions of your business depending on how they're asked. Here's what causes it and how to fix it.
What AI Search Means for Canadian Small Businesses
If you run a small business in Burlington, Hamilton, or Toronto, AI is already deciding whether to recommend you. Most Canadian SMBs have no idea what it's saying. Here's why, and what to do about it.
Why AI Still Won't Cite Your Business (After Fixing the Basics)
You've added schema, cleaned up your directories, and claimed your GBP. AI still won't recommend you. Here's what's actually going on.
AI Visibility for Accountants and Financial Advisors
When someone asks AI to recommend an accountant, trust signals matter as much as keywords. Here's how financial pros get cited — most aren't.
What To Do When AI Gets Your Business Wrong
AI models sometimes cite the wrong address, hours, or description for your business. Here's why it happens and how to correct the record.
AI Visibility for Real Estate Agents
Buyers and sellers are starting their agent search with AI. Real estate is one of the toughest visibility categories, and most agents are set up wrong.
How to Write an llms.txt File: Step-by-Step Guide
An llms.txt file gives AI models a direct brief on your business. Here's exactly what to write in each section — with real examples.
How AI Search Differs From Google for Local Businesses
Google ranks your website. AI search decides whether to name you. For local businesses, that's a completely different game — and most aren't playing it yet.
Structured Data Types That Improve AI Citations
LocalBusiness schema is just the foundation. These five additional schema types are what push businesses from mentioned to regularly cited by AI models.
AI Visibility for E-Commerce Businesses
E-commerce sites face a different AI visibility challenge than local businesses. Here's what drives product citations across ChatGPT, Perplexity, and Gemini.
How to Get Cited by Claude and Gemini
Claude and Gemini pull from different data sources than Perplexity or ChatGPT. Tactics don't fully carry over — here's what each actually needs.
AI Visibility for Law Firms and Legal Services
Legal queries are among the most common things people ask AI. Here's why most law firm websites get skipped — and what to fix to start getting cited.
AI Visibility for Restaurants and Food Service Businesses
Diners are asking AI chatbots where to eat. If your restaurant isn't cited, you're losing reservations before anyone ever finds your menu.
AI Visibility for Dentists and Healthcare Providers
Patients are asking AI chatbots to recommend dentists and doctors. Healthcare is harder to get cited in — but the fixes are structural, not expensive.
AI Visibility for Contractors and Home Service Businesses
Homeowners are asking AI chatbots to recommend contractors. If your business isn't cited, you're losing jobs before a phone ever rings. Here's how to fix that.
The Role of Backlinks in AI Search Visibility
Backlinks still matter for AI visibility — but not the way they matter for Google. The signal AI models care about isn't link juice. It's corroboration.
How to Audit Your Own AI Search Visibility
You can run a basic AI visibility audit yourself in under an hour. Here's exactly how to do it — and how to interpret what you find.
What Happens When You Block GPTBot
Blocking GPTBot in robots.txt is a common mistake that silently removes businesses from ChatGPT recommendations. Here's what it does and when it matters.
Schema Markup Mistakes That Kill AI Visibility
Most valid schema still fails AI models. Here's the specific implementation mistakes we see most often in Signal Check audits — and how to fix them.
Why Directory Listings Matter for AI Recommendations
AI models use directory listings as corroborating evidence. Inconsistent or incomplete profiles are a top reason businesses score low in audits.
Why Your Google Reviews Affect AI Recommendations
Google reviews do more than build social proof — AI models read them to understand what your business does, where you do it, and whether to recommend you.
The Difference Between an AI Mention and an AI Citation
Being mentioned by an AI and being cited by one are not the same thing. Here's what separates them, why it matters, and how to move from one to the other.
What Is Answer Engine Optimization (AEO)?
AEO is the practice of getting your business cited in AI-generated answers — not just ranked in search results. Here's what it means and how it works.
How Perplexity Decides Which Businesses to Cite
Perplexity runs a live web search on every query — no training lag, no cached knowledge. That changes how you get cited, and improvements can show up in days.
How Your Google Business Profile Affects AI Recommendations
Your Google Business Profile isn't just for Google — it's one of the primary sources AI models use when deciding whether to recommend your business.
How to Write Service Pages That AI Models Actually Cite
Most service pages are written for humans who already decided to hire someone. AI models need something different — here's how to get cited.
What Is AI Search Visibility — And Why Your Score Matters
When someone asks ChatGPT for a dentist in your city, it doesn't search Google. It pulls from what it knows — and most local businesses aren't there.
How ChatGPT Decides Which Businesses to Recommend
ChatGPT doesn't rank websites. It synthesizes from training data and retrieval — and recommends the businesses that left the clearest signal.
The llms.txt File: The Fastest Win for AI Visibility
llms.txt is a plain text file you drop at your domain root telling AI models what your business does, who you serve, and how to describe you.
Why Your Competitor Shows Up in AI Answers (And You Don't)
When a Signal Check finds a competitor cited instead of our client, there's almost always a clear structural reason — rarely about who's better.
LocalBusiness Schema: What AI Engines Actually Read
Schema markup is structured data that tells search engines, AI models, and knowledge graph crawlers what your business is and what it does.
A Burlington Designer Scored 3.2/10 — Here's Exactly What We Found
Carson Teal is a Burlington designer with a clean portfolio site, good word of mouth, and a near-zero AI visibility score. Here's what we found.