The key is trustworthy data and easy citability. For your business to be recommended by AI (ChatGPT, Gemini, Google AI Overviews), you need to get three fundamentals right:

  • Google Business Profile (GBP): Filled out 100% – AI treats it as the primary “source of truth” about your services.
  • Consistency (NAP): Identical name, address, and phone number everywhere online – errors cost you the crawler’s trust.
  • Content: Answers on your site phrased concisely enough that AI can “lift” them and cite them as a ready-made solution.

If you want to know more, read the full article.

In 2026, local search is no longer just a fight for positions on a list of links. It’s a fight for the algorithm’s trust. Customer decisions are increasingly made inside generative answers (Google AI Overviews, ChatGPT, Perplexity, Gemini).

GEO (Generative Engine Optimization) is the natural evolution – optimizing for AI citations and recommendations. It calls for the precision of a Takumi (匠) master – a craftsman who tends to every detail of the information shared online.

Imagine the difference between an encyclopedia’s table of contents and a trusted advisor. Classic SEO is the fight to get your company’s name into the table of contents under the right letter. The user still has to search, compare, and choose on their own. GEO is the process of building authority so that the advisor (AI) – when asked how to solve a problem – points to you specifically, saying: „I know an expert who is best at this.”

Glossary

Before we get into the mechanics of GEO, we need to define the fundamentals that determine the strength of your digital authority:

  • NAP (Name, Address, Phone): Identical name, address, and phone number across the web – a critical factor in building the algorithm’s trust in the business’s location.
  • Passage Indexing: Google’s ability to analyze and rank individual paragraphs of text as standalone answers to queries.
  • E-E-A-T: A credibility framework based on Experience, Expertise, Authoritativeness, and Trust – essential for building the Knowledge Graph.
  • Schema.org (Structured Data): A meta-layer of code that labels the information on your page, removing any guesswork for AI about what your business actually is.

The new paradigm: why SEO isn’t enough

Classic search engines work on the principle of Retrieval (finding information). Crawlers index pages and then order them by ranking. AI models work on the principle of Generation (producing answers). They don’t just “find” information – they “read” it, “understand” it, and synthesize it into new content.

SEO remains the foundation (without indexing, AI won’t even learn you exist), but GEO is the interpretive layer.

  • SEO: Answers the “I want to find” micro-moments.
  • GEO: Answers the “give me a ready-made solution” need.

Algorithms: the truth about Passage Indexing (understanding the AI “brain”)

Many companies wrongly optimize their pages for the MUM (Multitask Unified Model), treating it like a new PageRank. That’s a mistake – MUM is not a direct ranking factor.

The key to success in GEO is Passage Indexing.

Google no longer evaluates a page only as a whole. It can extract a single paragraph (passage), assess its substantive value, and use it to build an answer in an AI Overview.

Practical implementation:

Your page structure must be modular. Each heading and the paragraph that follows it should form a closed, logical whole that AI can “lift out” and cite without losing meaning.

Content architecture: formatting for synthesis

LLMs (Large Language Models) work predictively. They process highly logically structured content most easily. “Answer-First” content (the answer up front) has the highest citation rate.

Ready-to-use implementation templates:

  • Template A: Definitions (“Answer-first”)
    1. Question (H2): A clear intent signal (e.g., “What is implantology?”).
    2. Definition: 2-3 sentences of substance, stated “plainly.” This is the passage for Passage Indexing.
    3. Context: A local example (builds the geographic connection).
    4. FAQ: Structured data for the bot.
  • Template B: Instructions (“Step-by-step”)
    AI loves lists because they’re easy to turn into advice.
    1. Goal (H2): “How to [achieve goal] in [city]?”.
    2. Abstract: A summary for the impatient.
    3. List: 5-7 steps (bulleted lists are a priority for AI).
    4. Verification: “How do you know it worked?”.
  • Template C: Data (“Data-Interpretation”)
    Describe every fact in the pattern: Fact -> Interpretation -> Implication for the customer.

Technical “translation”: the role of Schema.org

To an AI, a website is a string of characters. Structured data (Schema.org / JSON-LD) is a meta-layer that unambiguously labels those characters. It tells the crawler: “This string of digits is a phone number,” not a tax ID.

  • Priority: Implement the LocalBusiness, OpeningHoursSpecification, and GeoCoordinates types.

These are the technical names for the digital labels that let crawlers understand your business unambiguously. Without them, AI has to guess who you are.

  1. LocalBusiness: Tells the crawler: „This is a real business, not a blog”. That’s how AI knows people can visit or call you.
  2. OpeningHours: Says: „We’re open”. This is crucial when someone asks an assistant for an „open dentist right now” – if the crawler can’t see it, it’ll skip you.
  3. GeoCoordinates: Your exact GPS in code. So when a customer searches for someone „near me”, AI knows with mathematical precision that you’re the closest.

Implementation: No coding required. All you need is an SEO plugin (e.g., RankMath) where you enter the data – it “translates” it into code for the crawlers on its own.

  • The llms.txt file: An experimental solution that Google does not officially honor at the moment (it uses robots.txt). Treat it as an add-on, not a foundation.

Building authority: E-E-A-T and the Knowledge Graph

AI builds a Knowledge Graph – a network of connections between entities. Links matter, but AI also analyzes Brand Mentions (references without a link) and E-E-A-T signals to assess credibility.

The new E-E-A-T standard for local businesses:

Google and AI models assess credibility through the lens of 4 pillars. Here’s how to implement them:

1. EXPERIENCE (First-hand experience)

Goal: Show AI that you have hands-on experience (“first-hand experience”), not just theory.

Actions:

  • Publish case studies with concrete numbers (e.g., “Cavity treated in 45 minutes”).
  • Post “Before and After” photos (in compliance with GDPR).
  • Add short customer video testimonials (30-60 seconds).
  • Publish an “Expert Journal” – observations from your everyday work.
    Metric: AI cites your case study as proof of effectiveness.

2. EXPERTISE (Specialist knowledge)

Goal: Prove to AI that you have formal knowledge and industry recognition.

Actions:

  • Take part in webinars and talks (as a speaker).
  • Publish guest articles on industry sites (Brand Mentions).
  • Showcase certificates and licenses (mark them up with Schema: EducationalOccupationalCredential).
  • Run an industry newsletter (a signal of consistency).
    Metric: “expert + specialty” keywords in the TOP 3.

3. AUTHORITATIVENESS (Local authority)

Goal: Be recognized as a leader in your region.

Actions:

  • Get mentions in local media (e.g., city portals).
  • Get cited in rankings (e.g., “Top 10 dentists in [City]”).
  • Sponsor local events.
  • Maintain a complete profile on LinkedIn and in industry directories.
    Metric: Presence in the Google Knowledge Panel; at least 10 citations per month.

4. TRUSTWORTHINESS

Goal: Demonstrate transparency, safety, and ethics.

Actions:

  • Publish a transparent price list on your site (marked up with price Schema).
  • Respond to every review within 48 hours.
  • Ensure full GDPR compliance (Privacy Policy, Cookies).
  • Introduce your team on the “About us” page (photos + bios).

Metric: GBP rating >= 4.5; no warnings in GSC.

The heart of local GEO: Google Business Profile (GBP)

For local queries, the Google Business Profile is the “source of truth” for AI. Our tests show that GBP data (hours, services) is cited as a priority.

Practical implementation:

  • NAP consistency: Name, Address, and Phone must be identical across all sources (GBP, website, directories). Inconsistencies lead to AI “hallucinations.”
  • Review management: AI reads the content of reviews. Fight for reviews that contain keywords (e.g., “great collision repair”), because those are what rank you for specific services.

Conversational, Voice, and Visual Search

These systems (ChatGPT, Google Assistant) don’t display a list of results – they generate a synthesis. 55% of users search for local services (“dentist near me”) by voice.

A. VOICE SEARCH

Strategy: Voice queries are longer and more natural.

Actions:

  1. Change your keywords: Instead of “optician Gdańsk”, target “where can I find a good optician open right now in Gdańsk”.
  2. Deploy FAQ Schema: Add JSON-LD code with the questions you hear at the front desk.
  3. JSON: Practical implementation: the No-Code method (Merkle generator)

Instead of editing code by hand, you can label your page for AI using the intuitive Merkle Schema Markup Generator. It lets you implement the Local Business standard precisely, without the risk of syntax errors.

  • Configuration (choosing the type): In the tool’s menu, select “Local Business”. This defines your site as a real economic entity in the eyes of the crawlers.
  • Data mapping: Fill in the fields according to the NAP consistency principle. Don’t forget the “Opening Hours” section, which determines your visibility in “open now” queries.
  • Location (GeoCoordinates): Enter exact GPS coordinates to give AI mathematical precision in “near me” queries.
  • Integration: Once the code is generated (shown in the screenshot below), copy it and paste it into your SEO plugin (e.g., RankMath).

Metric: Growth in mobile traffic at a low CTR (the nature of voice).

B. VISUAL SEARCH

Strategy: Patients search for visual results (“smile makeover”) in Google Lens and Pinterest.

Actions:

  1. Quality and description: High-resolution photos with descriptive alt text: “tooth implant before and after – dentist Gdynia”.
  2. Image Schema: Mark up procedure photos as ImageObject.
  3. Platforms: Pinterest (procedure photos) + Google Images.
    Metric: Traffic from Google Images and recognition in Google Lens.

An architecture of continuous improvement

AI algorithms, like a living organism, evolve continuously. Your presence in generative answers today is no guarantee for tomorrow.

Practical implementation:

  • Digital monitoring: Once a month, type your key customer queries into ChatGPT, Gemini, and Google AI Overviews.
  • Gap analysis: If AI recommended a competitor, ask: „Why?”. Do they have newer photos? A more precise price list? More reviews confirming a specific skill? Identify the missing element and fill it in the next cycle.

The future: the symbiosis of SEO and GEO

We’re not heading toward an “AI-only” world. We’re heading toward a hybrid era.

  • Don’t abandon SEO – it’s your foundation of data and technical structure.
  • Add a GEO layer – it’s your reputation, your readability to crawlers, and your ability to be cited.

If your brand has consistent data (NAP), enjoys users’ trust, and delivers content formatted for algorithms to “lift” – you have a real chance of becoming the first recommendation. Whether the customer asks traditional Google or turns to an AI assistant for help.

Next steps: your “Quick Wins”

To turn GEO theory into real visibility gains, start with the three highest-impact actions you can do in 15 minutes:

  • Consistency audit: Check whether the NAP in your site footer is identical to the data in your Google Business Profile (GBP).
  • Review optimization: Reply to your latest review, naturally weaving in the name of a service (e.g., “implant repair”), which helps AI categorize your offering better.
  • E-E-A-T update: Add one real photo of a completed service to your GBP, building proof of your experience (“first-hand experience”).

Dividing roles in the implementation process

RoleStrategic responsibility
Owner / ManagerProviding case studies and proof of experience (photos, video testimonials).
SEO Specialist / WebmasterImplementing Schema structured data (Local Business, FAQ) and optimizing load times.+1
Content CreatorWriting content in the “Answer-First” format (Template A) for Passage Indexing .

Sample implementation checklist: tools and integrations

A practical action plan for a local business owner.

PHASE 1: FOUNDATIONS (Weeks 1-2)

  • Google Business Profile (full opt): Complete 100% of the fields. Add team photos, a service price list, and a Q&A section.
    Time: 4-6h | Tool: business.google.com
  • Schema.org & Structured Data: Implement JSON-LD for LocalBusiness and GeoCoordinates.
    Time: 2-4h | Tool: Google Tag Manager / RankMath plugin

PHASE 2: CITATION ECOSYSTEM (Weeks 3-4)

  • Mention monitoring: Set up alerts for your company name. Track where people write about you without a link.
    Time: 30 min/week | Tool: Google Alerts / Brand24 / Spotlight
  • AI visibility check: See how ChatGPT and Perplexity “see” you.
    Time: 15 min | Tool: ChatRank (free) / manual tests

PHASE 3: CONTENT & VOICE/VISUAL (Weeks 5-8)

  • FAQ for Voice Search: Create a page with 20 customer questions + FAQ Schema.
    Time: 4h | Tool: CMS editor
  • Image optimization: Compress photos and add ALT text (“service + city”).
    Time: 4h | Tool: TinyPNG
  • Building E-E-A-T: Publish your first case study with data.
    Time: 3h/article

PHASE 4: MONITORING (Ongoing)

  • KPI dashboard: Once a month, check: traffic from maps, number of calls, citations in AI.
    Time: 30 min/month | Tool: Google Looker Studio / GSC

Quick-access tools table:

ToolCostUse
Google Tag ManagerFreeDeploying Schema without a developer
RankMath SEO~$39/yearComprehensive SEO, FAQ Schema, Local SEO
Spotlight / Brand24PaidBrand mention monitoring
ChatRank / ChatGPTFreemiumChecking AI visibility
Google Search ConsoleFreeAnalyzing search results

FAQ

No – it’s „SEO + GEO”, not „vs”. SEO is the foundation (indexing, visibility); GEO is the reputation layer (whether AI recommends you). Secure your SEO basics first, then add GEO.

– Google (SEO): Returns a list of results – the customer has to search and compare on their own

– ChatGPT (GEO): Returns a ready-made answer with a specific recommendation – „I recommend you because you have X experience”

55% of users search for local services by voice in AI assistants. If you’re not there, you lose customers.

Three fundamentals:

1. Google Business Profile – filled out 100% (hours, services, photos, price list)

2. NAP consistency – the same name, address, and phone everywhere (GBP, website, directories)

3. Citable content – short answers, so AI can lift them as a ready-made solution

Without this, AI either doesn’t know you exist or doesn’t trust the information.

– Phase 1 (Foundations): 1-2 weeks (GBP, Schema.org) = appearing in Google Overviews

– Phase 2 (Authority): 3-8 weeks (case studies, reviews, mentions) = recommendations in ChatGPT –

Full result: 3-6 months (consistent recommendations)

It’s faster than traditional SEO, but it requires consistency.

Google now extracts individual paragraphs from pages, evaluates them, and uses them in AI answers. The result: you don’t have to be #1 for the whole page – one well-written paragraph is enough.

In practice: every H2 heading + the text that follows must form a self-contained whole that AI can cite.

AI reads the content of reviews, not just the stars. A review with specific keywords is a ready-made citation.

Example:

– Weak: „Good, satisfied”

– Strong: „Great implant procedure, after just a month I can eat normally again”

Fight for reviews that mention specific services – they’re what rank you in AI.

– Tools: 39 PLN/year (RankMath) + free (Google Tag Manager, Search Console)

– Work time: 40-50 hours in the first 2 months

– After that: 30 min/month (routine)

You can delegate or outsource it (an SEO/GEO agency).

Four pillars of credibility:

PillarWhat to do
E (Experience)Publish case studies with data, „before and after” photos
E (Expertise)Certificates, guest articles, webinars
A (Authoritativeness)Media mentions, TOP 10 rankings
T (Trustworthiness)Transparent pricing, GDPR, responding to reviews

AI reads all of this – the stronger your E-E-A-T, the more often it recommends you.

No. All you need is a plugin (e.g., RankMath – ~39 PLN/year). You fill in a form, the plugin converts it into JSON-LD code for the bots.

Alternatives: Google Tag Manager (free) or online schema generators.

Respond professionally within 48 hours. AI reads your replies. You show that you care about quality and solve problems – that builds credibility.

Example: „If you were in pain, that’s a signal to us that something went wrong. Please get in touch – we want to make it right.”

55% search for local services by voice. Queries are longer and natural: – Text: „dentist Gdańsk” – Voice: „where can I find a good dentist open right now in Gdańsk”

What to do: Add FAQ Schema with the questions you hear from customers. AI will answer with your content.

Yes, if: 1. You’re indexed (SMO/SEO basics) 2. You have citable content (Answer-First format) 3. You have E-E-A-T signals (reviews, case studies)

This is the first gateway to AI visibility – without it, ChatGPT won’t know you either.

Every month: 1. Type into ChatGPT: „I’m looking for a good [service] in [city]” 2. Type the same into Google 3. Check whether you show up

If you’re missing: Analyze the competitor – why does AI recommend them? Better pricing? Better reviews? Fill the gaps.

Patients search for visual results (a smile, an implant, a makeover) on Pinterest and Google Lens.

What to do: – High-resolution photos (min. 2000px) – Descriptive ALT text: „smile makeover before and after – veneer – dentist Gdynia” – Pinterest (procedure photos)

Payoff: Traffic from Google Images + recognition in Lens.

All of them, 100%, but what mainly reaches AI is:

– Opening hours (Voice Search: „open now”)

– Services + prices (the decision is made in AI) – Photos (Visual Search)

– Reviews (the content of the review, not just the stars) – Q&A section (you answer questions – AI reads them)

If something is missing – AI passes you by.

Answer-First format:

QUESTION (H2): What is implantology?

ANSWER (up front, 2-3 sentences):
Implantology is the branch of dentistry concerned with placing artificial teeth
where a tooth has been lost.

LOCAL CONTEXT:
In Gdynia we’ve been placing implants for 25 years.

DETAILS (if needed)
An implant is a metal cone screwed in…

AI reads the beginning – if a ready answer is there, it’s happy to cite you.

A Brand Mention is a reference to your business without a link. AI reads them as a signal of local authority.

Payoff: You appear in rankings (e.g., „Top 10 dentists in Gdynia”), sites write about you – AI sees that you matter.

Action: Secure mentions in local media, rankings, and industry directories.

PHASE 1 (Weeks 1-2): FOUNDATIONS

1. Google Business Profile – 100% complete (4-6h)

2. Schema.org (LocalBusiness, GeoCoordinates, OpeningHours) – (2-4h)

3. NAP consistency – check everywhere

PHASE 2 (Weeks 3-4): CITATIONS

1. Google Alerts – brand monitoring

2. Visibility test in ChatGPT/Google (15 min)

PHASE 3 (Weeks 5-8): CONTENT & VOICE

1. FAQ Schema – 20 customer questions

2. Case study – first experience with data

3. ALT text – procedure photos

Organic. AI recommends you for credibility, not for money. But it takes consistency: – Reviews (real experiences) – Content (real knowledge) – Transparency (real data)

Advertising is being visible; GEO is being recommended. The second is stronger – people trust AI recommendations more than ads.

1. Voice Search dominance – less and less typing, more and more speaking

2. Visual Search – images instead of keywords

3. Real service – fast replies to messages, transparency

4. Personalization – AI knows what a specific patient wants (age, history, needs)

Plan: Get ready now. In a year everyone will be doing GEO – you’ll have a 6-12 month head start.

Sources:

  1. https://www.linkedin.com/pulse/ai-passage-indexing-how-google-ranks-specific-sections-ayub-ansary-o8n1c
  2. https://www.infidigit.com/blog/what-is-passage-indexing/
  3. https://www.searchenginejournal.com/ranking-factors/google-mum/
  4. https://www.getpassionfruit.com/blog/why-ai-citations-lean-on-the-top-10
  5. https://www.searchenginejournal.com/ahrefs-data-shows-brand-mentions-boost-ai-search-rankings/559938/
  6. https://en.wikipedia.org/wiki/Generative_engine_optimization
  7. https://searchengineland.com/guide/knowledge-graph
  8. https://www.wisecube.ai/blog/how-google-uses-knowledge-graphs/
  9. https://collingmedia.com/search-engine-optimization-seo/google-my-business-local-ranking-factors/
  10. https://www.linkedin.com/pulse/top-strategies-optimize-your-google-business-wmbic
  11. https://neotype.ai/zeroclick-searches/
  12. https://thedigitalbloom.com/learn/2025-organic-traffic-crisis-analysis-report/
  13. https://www.jellyfish.com/en-gb/training/blog/how-zero-click-searches-are-changing-seo-in-2025
  14. https://www.singlegrain.com/seo/e-e-a-t-strategies-that-guarantee-googles-trust-in-2025/
  15. https://www.wearetg.com/blog/google-eeat/
  16. https://www.semrush.com/blog/eeat/
  17. https://www.omnius.so/industry-updates/google-adds-llms-txt-to-docs-after-publicly-dismissing-it
  18. https://www.wix.com/studio/ai-search-lab/llms-txt-myths
  19. https://testrigor.com/blog/rag-vs-ai-agents/
  20. https://learn.microsoft.com/en-us/azure/search/retrieval-augmented-generation-overview
  21. https://www.signitysolutions.com/blog/rag-vs-traditional-search-engines
  22. https://wiserreview.com/blog/local-seo-statistics/
  23. https://www.nizamuddeen.com/community/terminology/mum/
  24. https://blog.google/products/search/generative-ai-google-search-may-2024/
  25. https://awebdigital.co/google-search-mum-seo-strategies/
  26. https://blog.google/products/search/generative-ai-search/
  27. https://amplewebsol.com/understanding-passage-indexing-and-its-impact-on-seo/
  28. https://www.yellowhead.com/blog/google-ranking-factors/
  29. https://static.googleusercontent.com/media/www.google.com/en/search/howsearchworks/google-about-AI-overviews.pdf
  30. https://upward-impact.com/seo/insights/uncategorized/passage-indexing-and-how-google-ranks-individual-paragraphs/
  31. https://www.seoquantum.com/en/blog/google-mum-new-algorithm-will-impact-seo-or-not
  32. https://www.techtarget.com/whatis/definition/Google-Search-Generative-Experience-SGE
  33. https://wiserank.co.uk/google-passage-indexing/
  34. https://www.botify.com/insight/what-are-google-ai-overviews
  35. https://www.sistrix.com/tutorials/google-knowledge-graph-data-how-can-it-help-improve-my-seo/
  36. https://searchengineland.com/what-is-generative-engine-optimization-geo-444418
  37. https://onclusive.com/en-gb/resources/blog/what-is-generative-engine-optimization-geo-guide/
  38. https://blackpugstudio.com/news/local-seo-in-2025-how-to-rank-in-google-maps-and-the-local-pack47a10c3a9ad2
  39. https://www.walkersands.com/about/blog/generative-engine-optimization-geo-what-to-know-in-2025/
  40. https://www.q-tech.org/blog/what-is-google-knowledge-graph-and-how-does-it-work/
  41. https://www.mapranks.com/2025/06/26/ai-impact-on-google-business-profile-2025/
  42. https://www.linkedin.com/pulse/what-generative-engine-optimization-geo-complete-guide-singh-vzkqc
  43. https://www.google.com/intl/en_us/search/howsearchworks/how-search-works/ranking-results
  44. https://web4pro.pl/czynniki-rankingowe-google-moja-firma-2025/
  45. https://www.bluehost.com/blog/what-is-llms-txt/
  46. https://wellows.com/blog/brand-mentions-vs-citation/
  47. https://statescoop.com/llmstxt-government-websites-ai/
  48. https://writesonic.com/blog/ai-brand-mentions-vs-ai-citations
  49. https://www.amraandelma.com/best-zero-click-search-statistics/
  50. https://llmstxt.org
  51. https://ahrefs.com/blog/ai-brand-visibility-correlations/
  52. https://click-vision.com/zero-click-search-statistics
  53. https://semai.ai/blogs/aeo-trends-2025-emerging-answer-engine-optimization-strategies-for-the-ai-era/
  54. https://www.theedigital.com/blog/blog-how-to-dominate-local-search-rankings-in-competitive-markets
  55. https://cxl.com/blog/answer-engine-optimization-aeo-the-comprehensive-guide-for-2025/
  56. https://www.searchenginejournal.com/competitor-analysis-local-seo-gain-an-edge/542070/
  57. https://totheweb.com/blog/beyond-seo-your-geo-checklist-mastering-content-creation-for-ai-search-engines/
  58. https://geneo.app/blog/future-proof-content-ai-search-strategy-2025/
  59. https://www.meilisearch.com/blog/semantic-search-vs-rag
  60. https://blacksmith.agency/resources/digital-marketing/best-local-seo-strategies/
  61. https://surferseo.com/blog/answer-engine-optimization/
  62. https://aws.amazon.com/what-is/retrieval-augmented-generation/
  63. https://almcorp.com/blog/local-seo-strategies-for-2026-the-essential-guide/
  64. https://geneo.app/blog/schema-markup-structured-data-best-practices-geo-ai-search-2025/
  65. https://www.get-spotlight.com/articles/top-10-tools-for-ai-search-content-writing-and-conversational-search-optimization-in-2026/
  66. https://www.smartsitesai.com/ai-seo-insights/smart-schema-strategies-structuring-local-websites-for-ai-visibility/
  67. https://digitalscouts.co/blog/conversational-ai-in-2025.-voice-search-and-chatbots-for-b2b
  68. https://www.seo-kreativ.de/en/blog/e-e-a-t-guide-for-more-trust-and-top-rankings/
  69. https://201creative.com/structured-data-ai-search/
  70. https://www.ematicsolutions.com/conversational-seo-trend-2026/
  71. https://backlinko.com/google-e-e-a-t
  72. https://www.digidop.com/blog/structured-data-secret-weapon-seo
  73. https://zapier.com/blog/best-ai-chatbot/
  74. https://www.airops.com/blog/local-business-schema-markup-example
  75. https://www.vezadigital.com/post/ai-seo-how-to-optimize-for-ai-search-agents