DEV Community: KDUAI

What AI search recommended this week (28 September 2026)

KDUAI — Mon, 28 Sep 2026 13:41:21 +0000

What AI search recommended this week

28 September 2026 · Based on 120 recorded test runs · AI Visibility Index

AI search engines do recommend local businesses, but the pattern is not fixed: the names change by city and by wording, and the same service can surface different leaders when the prompt changes slightly. In the recorded tests for house cleaners, Denver showed the strongest concentration around a single business, while Austin and Chicago were more split across several names.

What the data shows

For house cleaner in Austin, TX, the top name in 20 test runs was Eleven Friends of Grumpy Brush, which appeared in 95% of answers. Sparkly Maid Austin Cleaning Services followed at 90%, then Tidy Hands Cleaners at 65%, Maids and Moore at 55%, and DF CLEANING LLC at 25%.

When the query changed to house cleaner in Austin, TX?, the mix shifted. Sparkly Maid Austin Cleaning Services led at 85%, while MoreHands Maid Service and Tidy Hands Cleaners were each named in 75% of answers. Eleven Friends of Grumpy Brush fell to 65%, and Maids and Moore to 40%.

For house cleaner in Chicago, IL, the most named business across 20 runs was The Deep Cleaners at 85%. The next group was much lower: House Keep Up at 45%, Homeaglow at 35%, Sparkly Maid Chicago at 30%, and Sparkly Maid at 30%.

The Chicago variation with a question mark again changed the ordering. The Deep Cleaners still led, but dropped to 55%. Molly Maid rose to 45%, Sweep Home Chicago reached 40%, King of Maids also reached 40%, and Happy Home Cleaning Chicago appeared in 35% of answers.

Denver was the most decisive market in this set. For house cleaner in Denver, CO, Alpine Maids appeared in 100% of answers across 20 runs. MaidPro Denver followed at 80%, Merit Maids at 65%, and both Superb Maids Denver and The Maids at 50%.

The Denver question-mark variant was still highly concentrated, but the lead changed. MaidPro Denver was named in 95% of answers, Alpine Maids in 80%, The Maids in 75%, Merit Maids in 70%, and Kathy Clean in 35%.

Taken together, the tests show that AI search engines are not returning one universal local ranking. They are surfacing a small set of recurring businesses, but the order can shift with tiny prompt changes, and the lead business can change even when the query is nearly identical.

What this means if you’re a local business

The practical lesson is that AI visibility is now a citation-and-entity problem, not just a traditional search ranking problem. If a business wants to be named consistently, it needs clean crawler access, strong entity signals across the web, and citable answers on its own site and in third-party listings.

That means the business name, address, phone number, service area, and category language should match everywhere the model is likely to check. It also means the website should plainly describe what the company does, where it operates, and what makes it a relevant answer for the query. Reviews, directory profiles, and local mentions matter because AI systems often assemble recommendations from multiple sources rather than from one authoritative list.

The data here also suggests a simple testing habit: ask the same service query in more than one form. A small punctuation change was enough to alter which businesses appeared most often in Austin, Chicago, and Denver. If the prompt changes the answer, the business is not yet fully stable in AI search.

FAQ

Are these the only businesses AI search engines recommend?

No. These tests show the most frequently named businesses in a limited set of prompts and 20 runs per prompt. Other businesses may appear in other queries, locations, or AI systems.

Does a higher percentage mean the business is always first?

No. It means the business was named in that share of answers, not that it was always placed first or framed the same way. The key pattern is consistency across repeated tests.

All numbers above come from recorded test runs published in our open data index. Want your business tested? One-time diagnostic, no subscription.

Originally published at the AI Visibility Index.

Who does AI recommend for dentist in Austin, TX??

KDUAI — Mon, 28 Sep 2026 11:39:05 +0000

{"@context": "https://schema.org", "@type": "Dataset", "name": "AI recommendation frequency: dentist in Austin, TX?", "description": "Recorded AI search test results (5 runs) showing which businesses AI engines name for dentist in Austin, TX?.", "url": "https://myaiplay.github.io/ai-visibility-hub/index/dentist-in-austin-tx?/", "creator": {"@type": "Organization", "name": "AI Visibility Index"}, "temporalCoverage": "2026-08/..", "license": "https://creativecommons.org/licenses/by/4.0/"}

Who does AI recommend when asked for a dentist in Austin, TX??

Method: we asked AI search engines real buyer questions ("best dentist in Austin, TX?", "who should I hire for dentist services near Austin, TX??") across 5 recorded test runs. Below: which businesses were named, and how often.

Rank	Business	Named in % of answers
1	Waterloo Dental	100%
2	Holly Phillip DDS	100%
3	ATX Family Dental	80%
4	North Austin Dentistry	80%
5	Grove Austin Family Dentistry	80%
6	Austin Artistic Dental	60%
7	David J. Gordon II, DDS	60%
8	Tech Ridge Dental	60%
9	Toothbar	40%
10	Vesga Advanced Proodontics	20%

Based on 5 recorded runs. AI answers vary between runs — treat as directional, not a ranking.

Is your business on this list?

If you're a dentist in Austin, TX? and you're not named above, there's a reason — and it's findable. The usual causes: AI crawlers blocked, weak entity signals, or no citable answer on your site. Start with our five-minute self-check, or get the full one-time diagnostic with your top 3 blockers and a 12-step fix plan.

Originally published at the AI Visibility Index.

What AI search recommended this week (21 September 2026)

KDUAI — Mon, 21 Sep 2026 06:25:10 +0000

What AI search recommended this week

21 September 2026 · Based on 119 recorded test runs · AI Visibility Index

AI search engines are already acting like recommendation layers for local services, and the names they surface are fairly consistent within each city and query phrasing. In these recorded test runs, a small set of businesses dominated each category, with top names appearing in a majority of answers.

What the data shows

For house cleaner in Denver, CO, the most frequently named business was Alpine Maids, which appeared in 100% of answers across 20 test runs. The next most visible names were MaidPro Denver at 80%, Merit Maids at 65%, and both Superb Maids Denver and The Maids at 50%.

A small wording change shifted the pattern. For house cleaner in Denver, CO? the leader changed to MaidPro Denver at 95% of 20 test runs, while Alpine Maids fell to 80%. The Maids rose to 75%, Merit Maids to 70%, and Kathy Clean appeared in 35% of answers.

That same sensitivity shows up in Australia. For landscaper in Brisbane, Australia, Designer Image was named in 75% of answers across 20 test runs, followed by Human Nature Landscapes at 70%, Logan Lawn Mowing & Garden Care at 55%, BGL Services at 45%, and Mow on lawn solutions at 40%. But for landscaper in Brisbane, Australia?, Aboveground Landscapes and Human Nature Landscapes were tied at 80%, with Martin Brothers at 60%, Crivello Landscapes at 50%, and Brizscapes at 45%.

Sydney showed a different cluster again. For landscaper in Sydney, Australia?, NBG Landscapes led at 85% of 20 test runs, while Dr Garden reached 70%. Concept Landscaping, Love Landscapes, and Sydney Landscaping Pty Ltd each appeared in 55% of answers.

Austin’s cleaning market was also highly concentrated. For house cleaner in Austin, TX, Eleven Friends of Grumpy Brush was named in 95% of 19 test runs, and Sparkly Maid Austin Cleaning Services followed closely at 89%. Tidy Hands Cleaners appeared in 63% of answers, Maids and Moore in 53%, and Mystik Cleaning Service in 26%.

What stands out across the tests

The strongest pattern is concentration: AI search engines repeatedly recommend a short list rather than spreading attention evenly across many local businesses. In every query set here, the top two or three names captured most of the visible recommendations.

Another clear pattern is query phrasing sensitivity. The Denver and Brisbane examples show that adding a question mark changed which business ranked first and how the rest of the list was distributed. For local businesses, that means visibility is not just about being “in the market”; it also depends on how the question is framed.

The city-to-city differences are equally notable. Denver leaned toward Alpine Maids and MaidPro Denver; Austin centered on Eleven Friends of Grumpy Brush and Sparkly Maid Austin Cleaning Services; Brisbane split between Designer Image and Human Nature Landscapes in one formulation, then Aboveground Landscapes and Human Nature Landscapes in another; Sydney settled on NBG Landscapes.

What this means if you’re a local business

If an AI search engine is part of the customer journey, your business needs to be easy for systems to parse, trust, and cite. That means clear entity signals across your website, consistent business details across directories, and service pages that answer common local questions in plain language.

Businesses also need content that can be lifted into a direct answer. AI systems tend to favor names they can confidently associate with a place, a service, and a relevant query. Strong crawler access, structured service information, and third-party mentions all help reduce ambiguity.

The practical takeaway is simple: local visibility is no longer only about map packs and organic rankings. It is also about whether an AI system can confidently recommend your business when someone asks a local service question.

FAQ

Do these results show the single best local business in each city?

No. They show which businesses were most often named in these recorded AI search test runs for specific prompts.

Can a small wording change affect which business gets recommended?

Yes. The Denver and Brisbane results changed noticeably when a question mark was added, which shows how sensitive AI recommendations can be to query phrasing.

All numbers above come from recorded test runs published in our open data index. Want your business tested? One-time diagnostic, no subscription.

Originally published at the AI Visibility Index.

Who does AI recommend for dentist in Austin, TX?

KDUAI — Mon, 21 Sep 2026 05:22:48 +0000

{"@context": "https://schema.org", "@type": "Dataset", "name": "AI recommendation frequency: dentist in Austin, TX", "description": "Recorded AI search test results (4 runs) showing which businesses AI engines name for dentist in Austin, TX.", "url": "https://myaiplay.github.io/ai-visibility-hub/index/dentist-in-austin-tx/", "creator": {"@type": "Organization", "name": "AI Visibility Index"}, "temporalCoverage": "2026-08/..", "license": "https://creativecommons.org/licenses/by/4.0/"}

Who does AI recommend when asked for a dentist in Austin, TX?

Method: we asked AI search engines real buyer questions ("best dentist in Austin, TX", "who should I hire for dentist services near Austin, TX?") across 4 recorded test runs. Below: which businesses were named, and how often.

Rank	Business	Named in % of answers
1	Walden Dental	100%
2	Holly Phillip DDS	100%
3	South Austin Dental Associates	50%
4	Dr. Tor Gotun, DDS	50%
5	Toothbar	25%
6	Austin Dental Works	25%
7	Celebrate Dental & Braces South Austin	25%
8	AK Dental of Austin	25%

Based on 4 recorded runs. AI answers vary between runs — treat as directional, not a ranking.

Is your business on this list?

If you're a dentist in Austin, TX and you're not named above, there's a reason — and it's findable. The usual causes: AI crawlers blocked, weak entity signals, or no citable answer on your site. Start with our five-minute self-check, or get the full one-time diagnostic with your top 3 blockers and a 12-step fix plan.

Originally published at the AI Visibility Index.

What AI search recommended this week (14 September 2026)

KDUAI — Mon, 14 Sep 2026 06:25:35 +0000

What AI search recommended this week

14 September 2026 · Based on 114 recorded test runs · AI Visibility Index

AI search engines are already recommending a fairly consistent set of local landscapers, but the winners vary a lot by city and even by punctuation in the query. In the recorded test runs, some businesses appeared in a majority of answers, while others only surfaced in one phrasing and not the other.

The clearest pattern is that AI systems are not returning a single universal local leader; they are surfacing different businesses depending on the exact wording of the question. That makes local AI visibility less like traditional ranking and more like a mix of entity recognition, query interpretation, and answer selection.

What the data shows

Austin, TX

For the query “landscaper in Austin, TX?”, the most frequently named business was Southern Love Landscaping & Design, which appeared in 84% of answers across 19 test runs. Ground & Garden followed at 79%, then Top Choice Lawn Care at 63%, Austin Creative Landscaping at 47%, and Anderson Landscapes at 42%.

That result set suggests a relatively concentrated answer pattern, with two businesses clearly standing out above the rest. The top five were all named in a substantial share of responses, but the gap between the top pair and the lower entries was still meaningful.

Chicago, IL

Chicago produced one of the most striking examples of query sensitivity. For “landscaper in Chicago, IL”, Cityscape Landscape was named in 79% of answers, Patch Landscaping in 63%, Christy Webber Landscapes in 47%, and both Christy Webber Landscaping and Bruce Lawn Service in 26%, again across 19 test runs.

When the query changed to “landscaper in Chicago, IL?”, the distribution changed sharply. Christy Webber Landscapes reached 100% of answers, Mariani Landscape appeared in 79%, Topiarius in 53%, Cityscape Landscape LLC in 47%, and Chicago Specialty Gardens in 42%.

The difference between these two Chicago prompts is the strongest evidence in this dataset that small wording changes can shift which businesses AI search engines recommend. It also shows that a business may be visible under one phrasing and much less visible under another, even when the intent looks nearly identical.

Melbourne, Australia

Melbourne showed a similar pattern of query variation. For “landscaper in Melbourne, Australia”, Ian Barker Gardens was named in 68% of answers, Oneflare in 58%, Garden More Landscaping in 47%, Grassroots Landscaping & Maintenance in 37%, and Love It Landscaping in 32%.

For “landscaper in Melbourne, Australia?”, Bayside Landscaping dominated at 100%, followed by KD Landscapes at 68%, Ian Barker Gardens at 53%, Love It Landscaping at 53%, and Garden More Landscaping at 47%.

Here too, the presence or absence of a question mark reshaped the answer set. Ian Barker Gardens remained visible in both versions, but Bayside Landscaping jumped to the top only in the version with the question mark.

Phoenix, AZ

Phoenix was another market where a few names dominated the response set. For “landscaper in Phoenix, AZ”, Diamond Stone & Synthetic Grass was named in 74% of answers, BIG BOSS Landscape in 63%, Unwind Landscapes in 58%, Merit Landworks in 37%, and Franks Hardscape & Landscape Design LLC in 32%.

Unlike Chicago and Melbourne, Phoenix did not have a second query variant in the recorded data, so the main takeaway is concentration rather than phrasing sensitivity. The top three businesses were all named in more than half of answers, which indicates strong visibility within this test set.

What this means if you’re a local business

For local businesses, AI visibility now depends on more than just being findable in maps or search results. The systems need clear entity signals, crawlable pages, consistent business details, and answer-ready content that makes it easy to connect a query with a real company.

The practical implication is straightforward: businesses should make sure their name, address, phone number, service areas, and service categories are consistent across their website and major listings. They should also build pages that clearly explain what they do in plain language, because AI systems often prefer information that is easy to extract and cite.

Citable answers matter too. When AI systems generate local recommendations, they tend to favor businesses with strong, consistent entity footprints that can be summarized confidently. That means business pages, service pages, FAQ content, and reputable third-party mentions all help create a clearer signal.

FAQ

Why do the results change when the query wording changes slightly?

Because AI search engines may interpret nearly identical prompts differently and weight different business entities in the answer. The Chicago and Melbourne tests show that even a question mark can change which businesses are recommended.

Does being named in these answers mean a business is the best choice?

No. It only means the business was surfaced often in these recorded runs. These results measure visibility in AI answers, not service quality or customer satisfaction.

All numbers above come from recorded test runs published in our open data index. Want your business tested? One-time diagnostic, no subscription.

Originally published at the AI Visibility Index.

Why ChatGPT Doesn't Recommend Your Business

KDUAI — Mon, 14 Sep 2026 05:22:41 +0000

Why ChatGPT doesn't recommend your business (and how to check in 5 minutes)

16 August 2026 · AI Visibility Index

Direct answer: ChatGPT skips businesses for one of three reasons: (1) its crawlers can't read your site, (2) it can't confirm you're a specific, real business in a specific place, or (3) no page on the web gives it a short, sourced answer to quote. Check by asking ChatGPT, Google AI, and Perplexity "best [your category] in [your city]" and noting who gets named.

When a customer asks an AI engine who to hire, the answer isn't a ranking page — it's a sentence. "Here are three options in Denver." If your competitor is in that sentence and you're not, the sale is gone before your website had a chance.

Google rankings don't rescue you. Google AI uses Google's index, but ChatGPT leans heavily on Bing and its own browsing, and Perplexity has a separate retrieval stack. You can rank #1 on Google and still be absent from all three.

The five-minute check

Run the same two prompts on ChatGPT, Google AI (AI Overviews / AI Mode), and Perplexity:

best [your category] in [your city]
who should I hire for [your service] near [your city]?

Record three facts per engine: were you named, who was named instead, and whether the answer cited a source you control (your site, your Google Business Profile, a page you wrote).

Interpretation: if you're invisible on the recommendation prompt but present when someone asks "[your brand] reviews", you have a discovery problem, not a reputation problem. The model can talk about you when asked by name — it just won't volunteer you.

The three reasons you get skipped

1. Crawler access. A Disallow: / in robots.txt — or no explicit allow for GPTBot, ClaudeBot, PerplexityBot, and Google-Extended — means the model works from third-party pages about you, or nothing. Check https://yoursite.com/robots.txt before rewriting a word of copy. Our complete allow-list guide has the copy-paste config.

2. Weak entity. The model must know you're a specific business in a specific place — not a vague brand string. Consistent name/address/phone across directories, Organization or LocalBusiness schema on your homepage, a Google Business Profile, and listings in the directories your category actually uses are how you become a thing the model can name.

3. No citable answer. Pages that wander for 800 words before the point get summarised poorly. Lead each key page with a 75–150 word direct answer — who you serve, where, what you do, what it costs or how you work — then the evidence. Models prefer extractable blocks over vibes.

What to do this week

Unblock AI crawlers in robots.txt if they're blocked.
Add Organization/LocalBusiness JSON-LD matching your homepage.
Put a direct answer at the top of your main service page.
Re-run the three-engine prompt. Screenshot it. That's your baseline.

Frequently asked questions

Is this the same as SEO?

Related, but not identical. Google AI features use Google's index; ChatGPT relies on Bing and its own browsing; Perplexity has a separate stack. Ranking on Google does not guarantee you're the name in an AI answer.

Why would AI recommend a weaker competitor?

Models cite what they can retrieve as a clear entity with a short, sourced answer. A competitor with weaker service but cleaner schema, directory listings, and answer-first pages often wins the recommendation.

How often do AI answers change?

Constantly. No tool can guarantee identical results every run — which is why single-prompt screenshots mislead. Test each prompt multiple times and look at the pattern, not one answer.

Originally published at the AI Visibility Index.

What AI search recommended this week (07 September 2026)

KDUAI — Mon, 07 Sep 2026 06:25:51 +0000

What AI search recommended this week

07 September 2026 · Based on 108 recorded test runs · AI Visibility Index

AI search engines do not recommend every local business evenly. In these recorded test runs, a small set of landscapers consistently surfaced first, with different cities showing different “winners” and occasional name variations that split visibility across near-duplicates.

The clearest pattern is concentration: a few businesses dominate the answers, while several others appear often enough to matter but not enough to lead across every prompt variation. That makes local AI visibility look less like a broad directory listing and more like a shortlist that changes by city, phrasing, and naming consistency.

What the data shows

In Austin, the landscaper prompt produced a strong repeat leader. For “landscaper in Austin, TX?”, the most named business was Southern Love Landscaping & Design, appearing in 84% of answers across 19 test runs. Ground & Garden followed at 79%, Top Choice Lawn Care at 63%, Austin Creative Landscaping at 47%, and Anderson Landscapes at 42%.

Austin also showed how prompt wording can change the mix. For “landscaper in Austin, TX” across 18 test runs, Southern Love Landscaping & Design rose to 94%, while Anderson Landscape appeared in 89% of answers. Ground & Garden dropped to 39%, Top Choice Lawn Care to 22%, and Austin Pro Landscape to 17%. That spread suggests AI systems are not simply copying a fixed ranking; they are assembling recommendations from a moving set of signals.

Seattle showed a different pattern, with one business leading the way in the question form that included a question mark. For “landscaper in Seattle, WA?” across 18 test runs, Seattle Sustainable Landscapes appeared in 94% of answers, followed by Devonshire Landscapes Inc at 78%, Eagle Rock Landscaping Inc. at 67%, Landcrafters at 33%, and Seattle Gardening Co. at 22%.

For “landscaper in Seattle, WA” across 19 test runs, Devonshire Landscapes Inc became the most named business at 89%. New Life Rockeries & Retaining Walls appeared in 79% of answers, Urban Oasis Design & Construction LLC in 74%, and two similar entries — “Seattle Times Best in PNW nominee” at 26% and “Seattle Times Best in PNW Nominee” at 5% — showed how nearly identical names can fragment visibility when an AI system repeats or varies the phrasing.

Chicago was more concentrated at the top but still showed name fragmentation lower down. For “landscaper in Chicago, IL” across 17 test runs, Cityscape Landscape appeared in 76% of answers and Patch Landscaping in 65%. Christy Webber Landscapes appeared in 47%, while Christy Webber Landscaping showed up in 29% and Christy Webber in 24%. The split between those last three entries is a useful reminder that AI systems may treat similar business names as separate entities.

Denver showed a similar top-heavy pattern. For “landscaper in Denver, CO?” across 17 test runs, Lifescape Colorado appeared in 88% of answers. JR Landscapes LLC followed at 59%, Mile High Landscaping at 53%, Designs by Sundown at 41%, and Designs By Sundown at 35%. Here again, capitalization and naming variation appear to matter because the system surfaced two versions of the same-looking brand separately.

What this means if you’re a local business

The practical lesson is that AI search visibility is partly an entity problem, not just a ranking problem. If a business name appears in different forms across listings, reviews, directories, and press mentions, AI systems may split that authority across variants instead of confidently choosing one canonical name.

Local businesses that want to show up more often in AI answers should make their identity easy to crawl, easy to verify, and easy to cite. That means consistent business names, clear service descriptions, strong location signals, and pages that answer common customer questions in plain language.

It also helps to publish information that an AI can reuse directly: service areas, specialties, contact details, and proof points that are easy to quote without guessing. The goal is not just to rank in search results, but to become a reliable answer object that systems trust enough to name.

FAQ

Do AI search engines always recommend the same local business?

No. The test runs show that different prompt wording can change which business is named most often, even within the same city.

What is the biggest visibility risk for local businesses?

Inconsistent naming. When a business appears under multiple similar names, AI systems may split its visibility instead of reinforcing one clear entity.

All numbers above come from recorded test runs published in our open data index. Want your business tested? One-time diagnostic, no subscription.

Originally published at the AI Visibility Index.

What AI search recommended this week (31 August 2026)

KDUAI — Mon, 07 Sep 2026 05:22:42 +0000

What AI search recommended this week

31 August 2026 · Based on 78 recorded test runs · AI Visibility Index

AI search engines are already showing clear local preferences, and the names they repeat are not random. Across the recorded test runs, some businesses appeared in nearly every answer for the same query, while others surfaced only when the phrasing shifted slightly.

What the data shows

The strongest pattern in the dataset is concentration: a small set of businesses repeatedly dominates the answers for each local service query. For “dentist in Brisbane, Australia” across 13 test runs, four businesses were named in 100% of answers: North Brisbane Dental Clinic, Brisbane Dental, and Brisbane Dental and Denture Clinic; Dental Boutique Brisbane appeared in 85%, and SmileFresh Teeth Whitening Brisbane in 62%.

A near-identical topic but slightly different wording changes the mix. For “dentist in Brisbane, Australia?” across 13 test runs, Brisbane Dental was named in 100% of answers, Pure Dentistry in 77%, North Brisbane Dental Clinic in 54%, Brisbane Smiles in 38%, and St Lucia Dental in 31%. That contrast suggests that AI search results can be sensitive to small prompt changes even when the intent looks the same.

Sydney shows an even tighter top tier. For “dentist in Sydney, Australia?” across 13 test runs, Paramount Dental Sydney and Spa Dental Sydney CBD were each named in 100% of answers. Quay Dental Clinic appeared in 92%, while Pitt Street Dental Centre showed up in 62% and Pitt St Dental Centre in 38%.

The landscaping results are more distributed, but the leaders still stand out. For “landscaper in Austin, TX” across 13 test runs, Southern Love Landscaping & Design was named in 100% of answers, Anderson Landscape in 85%, Top Choice Lawn Care and Ground & Garden in 31% each, and Austin Pro Landscape in 23%.

When the query shifts to “landscaper in Austin, TX?”, the answer set reshuffles. Across 13 test runs, Ground & Garden, Top Choice Lawn Care, and Southern Love Landscaping & Design were each named in 77% of answers, Anderson Landscapes in 54%, and Austin Creative Landscaping in 46%. The exact business spelling also changes in the dataset, which is a reminder that AI answers may vary in how they surface and format local entities.

Denver is the most clustered landscaping market in this set. For “landscaper in Denver, CO” across 13 test runs, Mile High Landscaping, Designs by Sundown, and Designscapes Colorado were each named in 92% of answers. Little Foot Landscaping appeared in 62%, and The Fisherman LLC in 31%.

What this means if you’re a local business

If AI search engines are being used as a discovery layer, local businesses need to think like entities, not just websites. That means making sure the business name, address, phone number, service categories, and location signals are consistent across the web.

It also means crawler access matters. If a model or search system cannot reliably read your pages, structured data, or business profiles, you are less likely to be represented clearly in answer results.

Citable answers matter too. Pages that state services plainly, use unambiguous location references, and are easy to quote may be more likely to be surfaced in local recommendations. For multi-location brands, consistency across listings and site pages is especially important.

FAQ

Why do some businesses show up in every answer while others only appear sometimes?

The recorded runs suggest that AI search systems strongly prefer a small set of entities for each query. Small wording changes can also shift which businesses are surfaced.

Does this prove the AI is ranking the “best” local business?

No. The data only shows which businesses were named most often in the recorded test runs. It does not measure quality, conversion rate, or customer satisfaction.

All numbers above come from recorded test runs published in our open data index. Want your business tested? One-time diagnostic, no subscription.

Originally published at the AI Visibility Index.

What AI search recommended this week (24 August 2026)

KDUAI — Mon, 31 Aug 2026 06:11:22 +0000

What AI search recommended this week

24 August 2026 · Based on 47 recorded test runs · AI Visibility Index

AI search engines are already surfacing a small set of local businesses repeatedly, and the names vary by query phrasing even within the same city. In the recorded tests, some searches produced a single dominant recommendation, while others spread attention across several businesses and even non-business entities like Zocdoc or “specialty.”

What the data shows

The strongest pattern is that AI answers do not behave like a fixed local ranking. They are sensitive to query wording, and the same service plus city can produce a different set of recommended businesses.

For “dentist in Chicago, IL” across 8 test runs, the most named result was Chicago Loop Dentistry, appearing in 50% of answers. Three other clinics—William J. Millon DDS, Smile Science Chicago, and University Associates in Dentistry—each appeared in 38% of answers, while Zocdoc appeared in 25%. That mix suggests the answer set is broad enough to include both practices and a platform brand.

For “dentist in Chicago, IL?” also across 8 test runs, the top result changed completely: Millennium Park Dental appeared in 100% of answers. DiPilla Dentistry of Chicago followed at 88%, while The Magnificent Smile and Chicago Loop Dentistry each appeared in 50% of answers. Bite Club appeared in 38%. The presence of a question mark did not just alter emphasis; it changed the lead recommendation.

Melbourne showed a different pattern again. For “dentist in Melbourne, Australia” across 8 test runs, Smile Solutions appeared in 100% of answers. Pacific Smiles Dental Melbourne CBD appeared in 88%, Gorgeous Smiles in 75%, DENTL East Melbourne in 50%, and Core Dental Southbank in 38%. Here, the AI consistently surfaced several recognizable practices, but one name still stood out as universally named.

Phoenix split into two distinct answer sets. For “dentist in Phoenix, AZ” across 8 test runs, Jerome Riddle, DDS appeared in 100% of answers, with Biltmore Dental Center at 75% and Arizona Biltmore Dentistry at 62%. The remaining entries were Specialty and Insurance acceptance, each at 25%. That is a reminder that AI search sometimes recommends attributes or categories alongside businesses.

For “dentist in Phoenix, AZ?” across 8 test runs, Biltmore Dental Center became the dominant name at 100%. Musselman Dentistry appeared in 75% of answers, North Mountain Dentistry and The Biting Edge Family Dentistry each in 50%, and Downtown Phoenix Dental in 38%. Again, the query punctuation appears to shift which business the model treats as the safest answer to give.

Brisbane showed the clearest concentration of all. For “dentist in Brisbane, Australia” across 7 test runs, three businesses each appeared in 100% of answers: North Brisbane Dental Clinic, Brisbane Dental and Denture Clinic, and Brisbane Dental. Dental Boutique Brisbane appeared in 86% of answers, and SmileFresh Teeth Whitening Brisbane in 71%. That means the AI was not just recommending one option; it was repeatedly naming a small cluster of highly visible local businesses.

What this means if you’re a local business

AI search visibility looks less like classic blue-link SEO and more like entity selection. Businesses that show up consistently may have stronger crawler access, clearer entity signals, and more citable answers across the web.

If you want to be named more often, the practical goal is straightforward: make sure your business is easy for systems to identify, trust, and quote. That means consistent business names, strong location pages, structured contact details, and enough third-party references that an AI can confidently treat you as a real local entity.

It also helps to think beyond your own website. AI systems seem willing to draw from directories, review sites, and other sources that support a business’s identity. When those signals are clean and consistent, the business is more likely to be selected as an answer candidate.

FAQ

Why do the results change when the query wording changes?

The data shows that even small wording differences can shift which businesses are named most often. That suggests the model is interpreting the query slightly differently rather than relying on one stable local list.

Does being named in more answers mean the business is “winning” local AI search?

It means the business is appearing more often in these test runs, but not that it will always be selected everywhere. AI search visibility is query-sensitive, so the same business may perform differently across variants.

All numbers above come from recorded test runs published in our open data index. Want your business tested? One-time diagnostic, no subscription.

Originally published at the AI Visibility Index.

Schema Markup for AI Search: JSON-LD Guide

KDUAI — Mon, 31 Aug 2026 05:16:55 +0000

Schema markup for AI search: the JSON-LD that gets you cited

17 August 2026 · AI Visibility Index

Direct answer: schema markup (JSON-LD in your page HTML) is how you tell machines — unambiguously — "this is a business, named X, located at Y, offering Z". AI engines use these signals to build the entity understanding behind their recommendations. The two types that matter most for a local business are Organization/LocalBusiness on your homepage and FAQPage on Q&A content. Add them, keep them consistent with your visible content, and you become far easier for an AI to cite correctly.

Why AI engines care about schema

When ChatGPT or Perplexity decides who to name, it's really asking: which entities do I understand well enough to talk about confidently? A business described only in flowing marketing prose requires the model to guess. A business with clean JSON-LD hands it a form already filled in: legal name, address, phone, hours, services, ratings.

This is the same machinery behind Google's knowledge panels — but AI answer engines lean on it too, because structured data is the cheapest way for them to avoid making things up about you.

The four types worth your time

1. Organization or LocalBusiness (homepage). The foundation. Include name, address, phone, opening hours, service area, same-as links to your profiles:

<script type="application/ld+json">
{
  "@context": "https://schema.org",
  "@type": "LocalBusiness",
  "name": "Summit Plumbing",
  "address": {
    "@type": "PostalAddress",
    "addressLocality": "Denver",
    "addressRegion": "CO"
  },
  "telephone": "+1-303-555-0134",
  "url": "https://summitplumbing.example",
  "openingHours": "Mo-Su 00:00-24:00",
  "sameAs": [
    "https://www.facebook.com/summitplumbing",
    "https://www.google.com/maps/place/..."
  ]
}
</script>

2. FAQPage (Q&A sections). Question-and-answer blocks marked as FAQ are exactly the shape AI engines love to quote — a question a customer asks, with a direct answer attached.

3. Article (blog content). Author, date, headline — establishes freshness and attribution for your content.

4. Service or Product (offer pages). What you sell and at what price — lets AI answer "how much does X charge?" without guessing.

The rules that actually matter

Match the visible page. Markup describing content that isn't shown to visitors violates Google's guidelines and erodes trust signals everywhere.
Consistency beats completeness. Name, address, and phone identical across schema, your footer, your Google Business Profile, and directories. One inconsistency can split your entity in two.
Test it. Run your URL through Google's Rich Results Test and validator.schema.org after adding anything.

Not sure what shape your site is in? Our free AI Readiness Score checks your homepage's schema, indexability, and answer-first content in one pass.

Frequently asked questions

Is schema a ranking factor?

Not directly in classic SEO, and AI engines don't publish their weighting. What schema demonstrably does is improve entity understanding — which is the substrate recommendations are built from.

Do I need a developer to add JSON-LD?

Usually no. It's a paste-in script block; most CMSs have a field or plugin for it, and many SEO plugins generate Organization/LocalBusiness schema from a settings form.

Can wrong schema hurt me?

Yes — markup that contradicts your visible content (fake reviews, wrong addresses) can trigger spam policies and confuses AI engines. When in doubt, mark up less, accurately.

Originally published at the AI Visibility Index.

What Is llms.txt? Guide + Free Generator

KDUAI — Mon, 24 Aug 2026 08:30:03 +0000

What is llms.txt? The plain-English guide

17 August 2026 · AI Visibility Index

Direct answer: llms.txt is a plain-text file you place at yoursite.com/llms.txt that gives AI models a curated map of your site: who you are, what you do, and which pages matter most — in clean markdown, without navigation, ads, or boilerplate. Think of it as a briefing document for AI crawlers. It's an emerging convention, not a guarantee: adoption by the big AI providers is growing but not universal. It takes ten minutes to add and costs nothing.

What it actually looks like

A llms.txt file is markdown with a simple shape:

# Your Business Name

> One-paragraph summary: what you do, where, for whom.

## Services
- [Emergency plumbing](https://yoursite.com/emergency): 24/7 callout in Denver
- [Drain cleaning](https://yoursite.com/drains): fixed-price, same-day

## About
- [Pricing](https://yoursite.com/pricing)
- [Reviews](https://yoursite.com/reviews)

That's it. A title, a summary, and short lists of your most important links with one-line descriptions.

What it does — and what it doesn't

What it does: when an AI crawler or tool that respects the convention visits your site, it gets a clean, unambiguous briefing instead of guessing from your homepage's hero banner and cookie banner. For businesses whose sites are heavy on images and light on text, this is often the clearest description of the business that exists anywhere on the domain.

What it doesn't do: it's not a ranking switch. No major AI engine has promised that llms.txt alone gets you recommended. It works the same way good schema and clear copy work — it removes ambiguity, so the model can represent you correctly when it has reason to. Treat it as one layer in the stack, after crawler access and before content polish.

llms.txt vs robots.txt vs sitemap.xml

File	Audience	Purpose
robots.txt	Crawlers	Permission: what may be fetched
sitemap.xml	Search engines	Inventory: what exists
llms.txt	AI models	Briefing: what it all means

They complement each other. robots.txt says "you may enter"; llms.txt says "here's who we are once you're in."

How to make one in five minutes

Write your one-paragraph summary: business name, category, city, what you do, who for.
List your 5–10 most important pages with one-line descriptions.
Save as llms.txt in your site root so it serves at /llms.txt.
Verify: open https://yoursite.com/llms.txt in a browser and read what an AI would read.

We built a free generator that formats it for you: llms.txt Creator — fill in the fields, download the file, upload it to your site root.

Frequently asked questions

Do ChatGPT or Perplexity officially support llms.txt?

The convention is young and support is informal but growing — several AI crawlers and retrieval tools fetch it when present. Like early sitemaps, the cost of adoption is near zero and the direction of travel is clear.

Where exactly does the file go?

In your site's web root, so it's reachable at https://yourdomain.com/llms.txt. On most hosts that's the same folder as your homepage or your existing robots.txt.

Can llms.txt hurt my SEO?

No. It's a plain-text file ignored by anything that doesn't look for it. Google doesn't penalise extra text files, and it doesn't replace any SEO fundamentals.

Originally published at the AI Visibility Index.

ChatGPT Search Runs on Bing: Why It Matters

KDUAI — Mon, 24 Aug 2026 08:05:21 +0000

ChatGPT search runs on Bing — what that means for your business

17 August 2026 · AI Visibility Index

Direct answer: when ChatGPT searches the web to answer "who should I hire for X", it draws heavily on Bing's index — not Google's. A business can rank page one on Google and still be invisible to ChatGPT simply because Bing hasn't indexed it properly. The check takes two minutes: search site:yourdomain.com on Bing. No results means no ChatGPT citations from your own site.

The uncomfortable gap in most "AI SEO" advice

Most AI-visibility advice focuses on your website: schema, content structure, crawler access. All important. But there's a layer underneath that gets missed: the indexes the AI engines actually read from.

The three big consumer AI engines draw from different wells:

Engine	Primary retrieval source
Google AI (Overviews/AI Mode)	Google's own index
ChatGPT (search)	Bing's index + its own browsing
Perplexity	Its own retrieval stack + web search

Google's index and Bing's index are not the same thing. Bing crawls fewer pages, less often, for smaller sites. Plenty of local business sites that Google ranks happily are thin or absent in Bing — and that absence flows straight through to ChatGPT answers.

The two-minute check

Go to Bing and search: site:yourdomain.com
Count what comes back. Zero results — or just a bare homepage — is a red flag.
Do the same on Google for contrast. A big gap between the two is the story.

We automated this: the free Index Check tests your domain's Bing presence in one click.

How to fix a Bing gap

Bing Webmaster Tools (free): add and verify your site, then submit your sitemap. This alone fixes most cases within days.
IndexNow: Bing supports instant URL notification on publish — ask your developer or CMS plugin to enable it, so new pages get indexed in hours instead of weeks.
Check crawl blockers: confirm robots.txt allows Bingbot, and that your host or security layer isn't rate-limiting it.
Earn a few links from sites Bing already indexes well — local directories, industry associations, your suppliers. Bing leans harder on link discovery for small sites than Google does.

Why this compounds

Once you're properly in Bing, you don't just unlock ChatGPT search — Copilot, parts of Perplexity's retrieval, and a long tail of AI tools that licence Bing results all come along. One fix, many surfaces. And it's a fix most of your competitors haven't made, which is exactly the kind of edge that shows up in who gets named.

Frequently asked questions

If I'm in Google's index, am I in Bing's automatically?

No. They're separate crawlers with separate indexes and separate webmaster tools. Google inclusion tells you nothing about Bing — you have to check Bing directly.

How long does Bing indexing take after submitting a sitemap?

Typically days to a couple of weeks for an established domain; longer for brand-new ones. IndexNow can shorten this dramatically for new or updated pages.

Does Bing Webmaster Tools cost anything?

No — it's free, like Google Search Console. You can even import your Search Console data to skip manual verification.

Originally published at the AI Visibility Index.