April 7, 202617 min read

GEO for Insurance Companies: How Insurers Get Recommended by AI

How insurance companies can improve AI visibility in ChatGPT, Alice, Perplexity, and Google AI. YMYL specifics, expert content, insurance calculators, reviews, and Schema.org for insurers.

Vladislav Puchkov

Founder of GEO Scout, GEO optimization expert

According to monitoring data from geoscout.pro, insurance companies with transparent policy terms, online calculators, and active presence on niche platforms (Banki.ru, Yandex Maps, Sravni.ru) receive AI recommendations 3-5 times more often than companies with purely promotional landing pages. AI in insurance prefers specificity — coverage amounts, deductibles, payout timelines, calculation examples — not abstract promises of "reliable protection."

How AI handles insurance queries

When a user asks a neural network "which insurance company should I choose for auto insurance" or "where is the cheapest way to insure an apartment," the AI enters a zone of heightened responsibility. Insurance falls under the YMYL (Your Money or Your Life) category — a topic where an incorrect recommendation can lead to significant financial losses.

What this means in practice:

• AI adds disclaimers. Nearly every response to an insurance query includes "compare terms from multiple insurers" or "review the insurance policy terms carefully."
• Heightened trust threshold. Getting into recommendations requires more confirming signals than in standard niches: Central Bank data, agency ratings, real reviews.
• Preference for specialization. AI recommends insurers with strong positioning in a specific insurance type, not generic companies without focus.
• Filtering dubious claims. Promises like "we pay out in 1 day without documents" are ignored as unreliable.

Key numbers for insurance marketing:

• 51% of Russians use neural networks for decision-making, including choosing an insurer
• 88 million Alice users — many ask "Alice, where is the cheapest OSAGO" by voice
• 67% of policy buyers compare offers online before purchasing — AI is becoming part of this comparison
• The average insurance query to AI contains 20-35 words — describing the object, budget, and needs

Read more about general principles in the article on what GEO optimization is.

Types of insurance queries people ask AI

Potential customers turn to neural networks at different stages of choosing an insurer. Each query type requires its own content strategy.

Diversity of insurance queries — a niche characteristic

Unlike most niches, insurance queries to AI cover an extremely wide range of products: from compulsory OSAGO to niche insurance types. Each insurance type needs its own set of pages, FAQ, and calculators. A company covering 5-6 insurance types must create a separate content ecosystem for each.

YMYL specifics of insurance in AI

Insurance is a double YMYL zone: financial topics plus risks to property and health. AI approaches insurance recommendations with maximum caution.

Why AI does not recommend "just like that"

Neural networks are trained to avoid situations where their answers could lead to financial losses. In insurance, this means:

1. Requirement for multiple confirming sources. AI does not recommend a company based on one website. It seeks consensus: ratings, reviews, regulator data.
2. Priority of official data. The Central Bank of Russia publishes statistics on insurance companies — payout levels, complaint counts, sanctions. AI uses this data as a trust signal.
3. License filtering. If a company lacks an active Central Bank license or has had operational restrictions — AI factors this in.
4. Claims experience analysis. Not just "will they pay or not," but payout timelines, denial rates, service quality when an insured event occurs.

E-E-A-T for insurance companies

The E-E-A-T standard (Experience, Expertise, Authoritativeness, Trustworthiness) is as important in insurance as it is in medicine and law.

Optimizing an insurance website for AI

An insurance company's website is the primary source of information for AI. Page structure and content directly affect whether neural networks will recommend your company.

1. Insurance product pages with calculators

Each insurance type must have a dedicated page with a complete description. AI extracts specific data from these pages: coverages, exclusions, deductibles, application procedures.

Ideal product page structure:

• Product description — what it covers, who it suits
• Insurance terms — coverage amounts, deductibles, exclusions
• Online calculator — interactive policy cost calculation
• Calculation examples — 3-4 scenarios with specific amounts
• Application process — step by step
• What to do in an insured event — action algorithm
• FAQ — 15-20 questions about the product
• Comparison table with other products in your lineup

Pages you need to create:

2. Comparative coverage tables

AI loves comparative tables — they contain structured information that is easy to extract and compare.

Example: VHI program comparison

Such tables give AI specific data for comparison when answering "which VHI program to choose" queries.

3. FAQ for each insurance type

Detailed FAQs are a direct path to appearing in AI answers. Recommended format: 15-30 questions per insurance type.

Example questions for OSAGO:

• What determines the cost of an OSAGO policy in 2026?
• What to do if the at-fault driver has no OSAGO?
• How to calculate damage under OSAGO: European accident report or independent appraisal?
• Maximum OSAGO payout in 2026: property and health
• Can I get OSAGO online and is it legal?

Read more about FAQ markup for AI in the article on FAQ Schema for AI answers.

4. Schema.org for insurance companies

Structured markup is a critical GEO tool for insurers.

InsuranceAgency:

{
  "@type": "InsuranceAgency",
  "name": "AlfaStrakhovanie",
  "description": "Insurance company: OSAGO, KASKO, VHI, life and property insurance",
  "url": "https://www.alfastrah.ru",
  "telephone": "+7-800-333-0-999",
  "address": {
    "addressLocality": "Moscow",
    "streetAddress": "14 Shukhova St."
  },
  "areaServed": "RU",
  "hasOfferCatalog": {
    "@type": "OfferCatalog",
    "name": "Insurance products",
    "itemListElement": [
      {
        "@type": "Offer",
        "itemOffered": {
          "@type": "Service",
          "name": "OSAGO",
          "description": "Compulsory motor third-party liability insurance"
        }
      },
      {
        "@type": "Offer",
        "itemOffered": {
          "@type": "Service",
          "name": "KASKO",
          "description": "Comprehensive motor insurance against damage and theft"
        }
      }
    ]
  },
  "aggregateRating": {
    "ratingValue": "4.3",
    "reviewCount": "2840"
  }
}

FAQPage — on every product page.

Article — on expert publications with a named author-specialist.

5. E-E-A-T signals for insurance

AI verifies insurer expertise across several dimensions:

Trust and transparency:

• Central Bank license number prominently displayed (e.g., SI No. 2239)
• Link to the company page on the Central Bank website
• Financial reports — at least key indicators
• Information about shareholders and management
• Charter capital and reserves

Content expertise:

• Articles from insurance actuaries and experts
• Authors with stated qualifications and insurance industry experience
• References to legislation (OSAGO Law, insurance rules)
• Market reviews and analytics
• Analysis of changes in insurance legislation

Reviews and ratings as a factor in AI recommendations

Reviews are a critical signal for AI when selecting insurance companies. Neural networks analyze not only average ratings but also review substance, patterns, and frequency.

Where insurance companies need reviews

What kind of reviews AI values

Detailed reviews describing the insurance experience are the key factor. The ideal review for AI:

• Type of policy (OSAGO, KASKO, VHI, other)
• What happened (accident, illness, property damage)
• How quickly the insurer responded
• How the claims process went
• Final payout amount
• Service rating and recommendation

AI highlights recurring patterns. If 30+ reviews on Banki.ru mention "slow claim review" or "reduced payout" — that is a red flag for the neural network. If clients praise speed and transparency — it strengthens the recommendation.

Handling negative reviews

AI considers not only the presence of negative feedback but also the company's response:

• Company representative replies to complaints
• Speed of response to reviews
• Problem resolution or at least explanation of the position
• Rating dynamics over the last 6-12 months

External sources: ratings, reviews, Central Bank data

AI forms insurance company recommendations not only from the insurer's website. External sources that confirm or contradict company claims are critically important.

Reliability ratings

Central Bank data as a GEO factor

The Central Bank of Russia publishes detailed statistics on insurance companies that AI actively uses:

• Payout level — share of collected premiums paid out as claims
• Loss ratio — key indicator of financial stability
• Number of justified complaints — per 1,000 contracts
• Sanctions and restrictions — license suspension, enforcement orders
• Market share — company position in the market

Practical advice: Publish a "Transparency and Reliability" section on your website with data from Central Bank reports, accompanied by explanations. AI will value the company's willingness to disclose information.

Industry media and reviews

AI cites publications in industry and business media:

• Forbes Russia — insurance company rankings
• RBC — insurance market analytics
• Kommersant — reviews and insurance news
• Banki.ru — ratings and expert reviews
• Sravni.ru — insurance product comparisons
• Vedomosti — financial market analytics

Alice and Yandex: the priority channel for insurers

For insurance in Russia, Yandex with Alice is the key AI channel. Integration with Yandex Maps, Yandex Market, and financial services makes Alice the primary source of "which insurance to choose" recommendations.

How to optimize for Alice

1. Yandex Maps: complete business listing — all offices, hours, 4.5+ rating, interior photos
2. Yandex Business: up-to-date company data, product links
3. Reviews on Yandex: active review management — direct signal for Alice
4. OSAGO/KASKO calculators: Yandex Market integrates insurance calculators
5. Yandex Dzen: expert articles about insurance — Alice uses Dzen as a source

Read more about working with Yandex neural search in the article on how to check if Yandex neural search mentions your company.

ChatGPT and Perplexity

ChatGPT is used for more complex queries: comparing VHI programs, choosing life insurance types, analyzing policy terms. Perplexity cites sources directly — if your company is mentioned on Banki.ru or in business media, Perplexity will show this to the user. Read more about provider differences in ChatGPT vs Claude vs Gemini: who gets recommended.

Practical 30-day plan for an insurance company

A step-by-step plan for launching a GEO strategy at an insurance company.

Week 1: Audit and foundation

1. Check current AI visibility across 15-20 insurance queries in 5-6 neural networks via geoscout.pro
2. Conduct a GEO site audit — check Schema.org, robots.txt, page load speed, page structure
3. Verify the Central Bank license is present and correct on the website
4. Ensure robots.txt does not block AI bots (PerplexityBot, ChatGPT-User, GPTBot, ClaudeBot)
5. Verify data on Yandex Maps, Google Maps, Banki.ru, Sravni.ru

Week 2: Technical optimization

1. Implement Schema.org InsuranceAgency on the homepage and "About" page
2. Add FAQPage markup to all product pages (OSAGO, KASKO, VHI)
3. Create or update policy cost calculators
4. Add structured data for branches and offices (LocalBusiness)
5. Optimize page load speed — technical checklist

Week 3: Content foundation

1. Create product pages for each insurance type with complete term descriptions
2. Add comparative coverage tables (minimum for VHI and KASKO)
3. Write FAQs — 15-20 questions per product
4. Publish 5-7 expert blog articles: "How to choose KASKO in 2026," "OSAGO vs KASKO: what you need to know," "VHI for employees: an HR guide"
5. Add policy cost calculation examples for different scenarios

Week 4: External presence and monitoring

1. Intensify review management on Banki.ru, Yandex Maps, Sravni.ru
2. Set up monitoring of 20-30 insurance prompts via GEO Scout
3. Add competitors to monitoring (Sberbank Insurance, AlfaStrakhovanie, Ingosstrakh, Tinkoff Insurance, RESO-Garantia, VSK)
4. Create a 2-month content plan based on monitoring data
5. Set up Share of Voice tracking for key insurance types

Prompts for monitoring insurance queries

Prompt templates

Choosing an insurer:

• "Best insurance company for OSAGO in 2026"
• "Insurance company payout ratings for KASKO"
• "Which insurance to choose for VHI: reviews"

Comparison:

• "Sberbank Insurance or AlfaStrakhovanie — which has better KASKO?"
• "Ingosstrakh or RESO-Garantia — OSAGO comparison"
• "Tinkoff Insurance or VSK — which to choose for VHI?"

Cost and calculation:

• "How much does OSAGO cost for a Hyundai Solaris 2022 in Moscow?"
• "KASKO cost for a BMW X3 — what factors affect the price?"
• "VHI price for a 50-employee company in St. Petersburg"

Situational:

• "What to do if the insurance company is delaying my KASKO payout?"
• "How to file an OSAGO insurance claim after an accident — step by step"
• "Travel insurance for Turkey: what coverage is essential?"

Monitoring these prompts manually across 10 AI providers is unrealistic. GEO Scout enables automated daily monitoring. The Command Center generates a prioritized action plan: which pages to improve, what content to create, where to strengthen external presence.

GEO optimization checklist for insurance companies

E-E-A-T and trust:

• Central Bank license number displayed on the website with a link to the regulator's page
• Key financial indicators published (charter capital, reserves)
• Each expert article has a named author with qualifications
• Transparent insurance terms on every product page
• Information about management and shareholders

Product pages:

• Separate page for each insurance type (OSAGO, KASKO, VHI, life, travel, property)
• Online policy cost calculator on each page
• Calculation examples — minimum 3 scenarios per product
• Comparative coverage tables
• Step-by-step policy application instructions
• Action algorithm for insured events

Technical foundation:

• Schema.org InsuranceAgency markup on the homepage
• FAQPage markup on all product pages
• Article markup on expert publications
• robots.txt allows access to AI bots
• Website loads in under 3 seconds

Content:

• 15-20 expert articles on key insurance topics
• FAQ — 15-30 questions per insurance type
• Links to legislation and insurance rules
• Content is current for 2026 (tariffs, laws, terms)

External presence:

• Complete profile on Banki.ru with current reviews
• All office listings on Yandex Maps with 4.5+ rating
• Profile on Sravni.ru
• Reviews on 3+ platforms with detailed experience descriptions
• Presence in ratings (Expert RA, NRA)

Monitoring:

• 20-30 insurance prompts on daily monitoring
• Tracking 5-7 competing insurance companies in AI responses
• Share of Voice analysis by insurance type
• Monitoring mention sentiment across neural networks
• Content plan adjusted based on monitoring data