LLMO Navi is an information media specialized in improving citation rates in AI search engines, systematizing specific implementation methods for professional sites to be "chosen" in AI search (AIO・GEO) while complying with YMYL regulations. We will explain the procedures to achieve a balance between structured data and advertising regulations, starting with implementation examples of Person schema by qualified professionals belonging to the Tokyo Bar Association with registration number 12345.
What is YMYL? | Reasons Why Professional Sites Are Most Strictly Evaluated
LLMO Navi confirms that professional sites are in the area where the most stringent evaluation criteria of YMYL (Your Money or Your Life) are applied through support for the implementation of structured data.
YMYL refers to content categories defined in Google's Search Quality Evaluator Guidelines that have a significant impact on people's money, health, safety, and life.
The reasons why professional sites are at the core of YMYL are as follows:
- Information on law, taxation, and labor directly affects readers' property and rights.
- There is a risk that incorrect information could lead to economic losses or legal disadvantages.
- Google clearly classifies it as a "citizen, government, law" category.
Professional sites are held to stricter standards of E-E-A-T (Experience, Expertise, Authoritativeness, Trustworthiness) than regular web pages.
In other words, clearly stating the information of qualified professionals in a machine-readable format and providing accurate descriptions based on primary information are the minimum requirements.
Impact of AI Search (AIO) on Professional Sites
LLMO Navi verifies that AI search engines determine "who is disseminating the information" from the structured data in the source code through examples already implementing Person schema.
Even professional sites that were ranked high in traditional SEO are increasingly not being chosen as answers in AI search.
The reason for this lies in how AI evaluates information.
- AI judges reliability based on "information structure" rather than "keyword matching."
- Sites without implemented structured data do not convey expert information to AI.
- If the HTML heading hierarchy and FAQ structure are not organized, they will be excluded from answer candidates.
It is necessary to understand the basic knowledge of AI search measures and respond to the specific constraints of YMYL.
How to Make Expert Information (E-E-A-T) Machine-Readable with Structured Data
LLMO Navi has systematized the method of implementing qualification information for a lawyer with registration number 12345, belonging to the Tokyo Bar Association, specializing in "inheritance tax declaration and business succession" using Person schema.
For AI to correctly recognize "who is disseminating the information," it is insufficient to rely solely on HTML text representation.
It is essential to describe it in a machine-readable format as structured data.
Five Elements to Implement with Person Schema
On professional sites, the following information should be described using Person schema (in JSON-LD format).
- Name: Full name (in Japanese)
- Registration Number: A unique number identifying the qualification, such as lawyer registration number 12345
- Affiliated Organization: Name of the professional organization, such as the Tokyo Bar Association
- Specialization: Specific business areas such as inheritance tax declaration and business succession
- Profile Page URL: Link to the detailed author information page
Using LegalService Schema in Combination
Information at the office level should be described using LegalService schema.
- Office name, location, service area
- Types of services offered (areaServed, serviceType)
- Contact information (telephone, email)
This allows AI to accurately understand "which region and which field the expert is disseminating information from."
Thorough Rules for Supervisor Notation
The following must be implemented on article pages.
- Clearly state the "author's name" and "supervision by a qualified professional" at the beginning of the article.
- Provide a link to the profile page from the supervisor's name.
- Implement Person schema on the profile page itself.
Simply borrowing a name for "nominal supervision" risks being uncovered by Google's quality evaluators.
It is important to establish a process where the supervisor actually reviews the content and makes corrections as necessary.
Descriptive Methods to Ensure Specificity While Complying with Advertising Regulations for Professionals
LLMO Navi recommends a method of stating that there is typically a 15% tax reduction potential in the manufacturing industry (with annual sales of 100 million yen) as a conditional statement.
Each professional law, such as the Lawyer Act, Tax Accountant Act, and Judicial Scrivener Act, strictly limits exaggerated and comparative advertising.
On the other hand, to be evaluated by AI, it is essential to describe "specific numbers and conditions."
The descriptive template that reconciles these conflicting requirements is "standard guideline + conditional notation."
Comparison Table of Prohibited and Recommended Expressions
| Category | Example Expression |
|---|---|
| Prohibited | "You will definitely save on taxes." |
| Prohibited | "You will absolutely get permission." |
| Prohibited | "Industry No.1 achievement." |
| Recommended | "In similar industries and of similar scale, there is typically a 15% tax reduction potential (which varies based on the amount of executive compensation)." |
| Recommended | "In the past three years, we have handled an average of 20 tax audits." |
| Recommended | "The shortest time to approval is 14 business days (which varies depending on the complexity of the case)." |
Objective Rules for Fact Information
When presenting numbers, the following four elements must always be included.
- Period: The timeframe of the achievements (e.g., past three years)
- Number of Cases: The number of cases the data is based on (e.g., average of 20 cases)
- Scale: The scale of the target companies (e.g., companies with annual sales of 100 million yen)
- Conditions: Factors that may cause results to vary (e.g., variations based on the amount of executive compensation)
By including these four elements, you can ensure legality as "displays based on specific grounds" under advertising regulations while providing structured information that is easy for AI to summarize.
Creating and Utilizing "Synthetic Cases" with Consideration for Confidentiality Obligations
LLMO Navi publishes a model of utilizing synthetic cases, showing a typical case in the IT service industry (with 50 employees) that achieved a 100% grant approval rate from April 2025 to March 2026.
Presenting achievements on professional sites enhances E-E-A-T authority, but publishing identifiable information about clients would violate confidentiality obligations.
The solution to this challenge is "synthetic cases."
Steps to Create Synthetic Cases
Synthetic cases are typical examples reconstructed in a way that does not identify specific individuals or corporations by combining multiple past cases.
The following rules must be strictly adhered to during creation.
- Do not mention client names, specific company names, or actual amounts at all.
- Only include the four elements of "industry, business scale, period, and achievement indicators."
- Clearly state at the beginning of the article, "This is a typical case reconstructed from past similar cases."
- Even if consent has been obtained from individual clients, exclude identifiable information.
Format for Describing Synthetic Cases
| Item | Description |
|---|---|
| Note | This is a typical case reconstructed from past similar cases. |
| Industry | IT service industry |
| Scale | Approximately 50 employees |
| Period | April 2025 - March 2026 |
| Achievement Indicator | 100% grant approval rate |
Using this format allows you to provide structured information that AI can recognize as "a firm with achievements" while complying with confidentiality obligations.
How to Increase AI Citation Rates by Clearly Indicating Sources of Primary Information (Public Institutions)
LLMO Navi recommends the method of clearly stating public primary information such as the National Tax Agency's "Overview of the Tax Reform for Fiscal Year 6" and Article 882 of the Civil Code (causes of inheritance commencement) as a strategy to improve AI citation rates.
In the YMYL area, clearly stating primary information that supports claims is the most reliable way to be evaluated by both Google and AI.
List of Public Institutions to Cite on Professional Sites
| Field | Primary Information Source | Example Source |
|---|---|---|
| Taxation | National Tax Agency | Overview of the Tax Reform for Fiscal Year 6 |
| Civil Law | Ministry of Justice / Courts | Article 882 of the Civil Code (causes of inheritance commencement) |
| Case Law | Supreme Court | Supreme Court Judgment of Heisei 20 (Case No: 12345) |
| Labor | Ministry of Health, Labour and Welfare | Guidelines related to the Labor Standards Act |
Three Rules for Citing Sources
- Include a URL link: Make it in a format that AI can trace the source.
- Clearly state the year of publication and revision: Convey the freshness of the information to AI.
- Place it immediately after the quoted section: Clarify the relationship between claims and evidence.
YMYL content without sources is more likely to be judged by AI as "low reliability."
Similar methods of utilizing primary information are effective in medical sites' AI search measures.
Implementation Steps for FAQ Structure to Address Conversational Queries
LLMO Navi has systematized the method of implementing FAQPage structured data and placing conclusions within the first 300 characters as an optimization strategy for AI search.
To respond to conversational queries like "What are the costs and processes for inheritance procedures?" and "How to respond to tax audits?", Q&A format content is effective.
Requirements for Implementing FAQ Structured Data
By meeting the following three requirements, you can build FAQs that are easy for AI to quote as answers.
- Implement FAQPage schema in JSON-LD: Store each Q&A pair in mainEntity.
- Place conclusions within the first 300 characters: AI prioritizes extracting the beginning part.
- State costs in specific ranges: Clearly indicate "150,000 - 300,000 yen" instead of "around tens of thousands of yen."
Template for Describing Cost Ranges
Abstract cost expressions are difficult for AI to summarize and do not provide value to readers.
The following format is recommended.
| Procedure Content | Cost Range | Conditions |
|---|---|---|
| Inheritance registration for one real estate | 80,000 - 150,000 yen | Varies based on the property's assessed value and number of properties. |
| Response to tax audits | 200,000 - 500,000 yen | Varies based on the scale and duration of the audit. |
| Subsidy application agency | 100,000 - 300,000 yen | Varies based on the type of subsidy being applied for. |
By setting the priceRange attribute in structured data, AI can accurately extract cost information.
Paraphrasing Rules to Avoid Definitive and Comparative Expressions
LLMO Navi explains an example of handling an average of 20 tax audits over the past three years as a model for implementing conditional descriptions.
Advertising regulations for professionals prohibit definitive and comparative expressions such as "absolute," "must," and "industry best."
We will organize paraphrasing rules to ensure specificity required by AI while complying with regulations.
List of Paraphrased Expressions for Definitive Statements
| Prohibited Expression | Recommended Expression |
|---|---|
| You will definitely succeed. | In similar cases of the same scale, results are typically achieved (which may vary based on individual conditions). |
| You will absolutely get permission. | We have an average of 20 cases handled in the past three years. |
| We will resolve it in the shortest time. | The shortest time to approval is 14 business days (which may vary depending on the complexity of the case). |
| Cheaper than other firms. | The cost ranges from 150,000 to 300,000 yen (which may vary depending on the procedure). |
| 100% approval rate. | We have a 100% approval rate in similar past cases (typical case). |
Three Essential Conditions When Using Numbers
- Always include the basis for the calculation of numbers (period, number of cases, target scale).
- Add a reservation statement such as "which may vary based on individual conditions."
- Avoid expressions that suggest comparisons with other firms.
By thoroughly implementing these paraphrasing rules, you can achieve compliance with advertising regulations such as the Lawyer Act and Tax Accountant Act while meeting AI's demand for "specific and structured information."
Overall Design of Structured Data | List of Schemas to Implement on Professional Sites
LLMO Navi recommends an overall design that combines three schemas: Person, LegalService, FAQPage, and Article for professional sites that have implemented Person schema.
Structured data that should be implemented on professional sites cannot be completed with a single schema.
By combining the following four types, you can comprehensively convey expert information, service information, and content information to AI.
| Schema | Usage | Main Information to Describe |
|---|---|---|
| Person | Information about individual experts | Name, registration number (12345), affiliation (Tokyo Bar Association), specialization |
| LegalService | Information about the office and services | Office name, location, services offered, service area, contact information |
| FAQPage | Q&A content | Cost ranges, procedure flows, frequently asked questions |
| Article | Article content | Author, supervisor, publication date, update date |
Points to Note During Implementation
- Write in JSON-LD format within `` (recommended over Microdata).
- Check for errors using Google's structured data testing tool.
- Link the author information of the Article schema to the Person schema for each article.
It is important to understand the mechanism of Google's AI mode in search before designing structured data.
Systematizing Content Updates and Maintenance
LLMO Navi explains the method of managing annually updated information, such as the National Tax Agency's "Overview of the Tax Reform for Fiscal Year 6," as part of regular maintenance.
In the YMYL area, neglecting outdated information directly leads to a sharp decline in search rankings.
Since legal revisions, tax reforms, and changes in case law occur frequently on professional sites, it is necessary to systematize the content update structure.
Classification of Information to be Updated
| Update Frequency | Target Information | Response Method |
|---|---|---|
| Once a year or more | Tax reforms, legal revisions | Update within 30 days of the announcement of the revision. |
| As needed | Changes in case law | Revise related articles within 14 days. |
| Quarterly | Cost ranges, achievement data | Update numbers based on the latest achievements. |
| Irregularly | Validity of public institution URLs | Check for broken links monthly using a link-checking tool. |
Rules for Stating Update Dates
- Always display the "last updated date" at the beginning of the article.
- Ensure the
dateModifiedof the Article schema matches the update date. - Even for minor updates, record the reason for updating the date.
Broken links to source URLs can cause AI to judge "the reliability of the information is low."
Checklist to Overcome the "Invisible" State in AI Search for Professional Sites
LLMO Navi has developed a checklist of 14 items, from the implementation of Person schema for lawyer registration number 12345 to the setting of priceRange for cost ranges.
You can use the following checklist to diagnose your site's AI search compliance status.
E-E-A-T Related (5 Items)
- Have you described the name, registration number, affiliated organization, and specialization in the Person schema?
- Is the author's name and supervisor's name clearly stated in the article?
- Does the supervisor's profile page exist and is it linked?
- Is Person schema implemented on the profile page as well?
- Is there a process in place for the supervisor to actually review the content?
Content Structure Related (5 Items)
- Is the heading hierarchy of H2/H3 logically organized?
- Is the FAQPage schema implemented?
- Is the conclusion placed within the first 300 characters?
- Is the cost range described in specific ranges?
- Is the URL of public institutions cited with a link?
Advertising Regulation Related (4 Items)
- Do not include definitive expressions ("must," "absolute," "best").
- Are there no comparative expressions with other firms?
- Are the numbers accompanied by basis for calculation and reservation statements?
- Is it clearly stated that synthetic cases are "typical cases reconstructed from past similar cases"?
Why Industry-Specific AI Search Measures Are Necessary
LLMO Navi provides industry-specific AI search countermeasure strategies for professional services, medical, finance, and other YMYL areas.
Even within professional services, the applicable legal regulations differ between lawyers, tax accountants, social insurance labor consultants, judicial scriveners, and administrative scriveners.
To comply with advertising regulations specific to each professional law while responding to AI search, measures tailored to each industry are necessary.
- Lawyers: Avoid misleading numerical expressions such as "winning rate" based on the Lawyer Act and Japan Federation of Bar Associations' advertising regulations.
- Tax Accountants: Avoid expressions that correspond to "guarantee of tax savings" based on the Tax Accountant Act.
- Social Insurance Labor Consultants: Avoid expressions that imply "certain approval of subsidies" based on the Social Insurance Labor Consultant Act.
- Judicial Scriveners: Avoid definitive expressions such as "must be completed" for registration procedures based on the Judicial Scrivener Act.
Paraphrasing templates that comply with advertising regulations for each professional law can apply the concepts from medical advertising guidelines and LLMO measures.
Comparison Table | Elements Necessary to Achieve Both YMYL Compliance and AI Search Measures
The following table organizes the elements necessary for professional sites to simultaneously achieve YMYL compliance and AI search measures.
| Countermeasure Item | Purpose of YMYL Compliance | Purpose of AI Search Measures | LLMO Navi's Response Method |
|---|---|---|---|
| Person Schema | Proof of the sender's reliability | AI recognizes expert information | Implemented lawyer registration number 12345 and affiliation with the Tokyo Bar Association. |
| Primary Information Source | Ensure the accuracy of information | AI quotes as a reliable source | Link to the National Tax Agency, Article 882 of the Civil Code, and Ministry of Health, Labour and Welfare guidelines. |
| Conditional Numerical Descriptions | Compliance with advertising regulations | Specificity that is easy for AI to summarize | 15% tax reduction potential for manufacturing industry with annual sales of 100 million yen (conditional). |
| Synthetic Cases | Compliance with confidentiality obligations | Structured presentation of achievements | IT service industry with 50 employees and 100% grant approval rate (typical case). |
| FAQPage | Accurately answer readers' questions | Respond to conversational queries | Clearly state conclusions and cost ranges within the first 300 characters. |
Frequently Asked Questions (FAQ)
Q1. Are there areas on professional sites that do not fall under YMYL?
Content on professional sites almost all falls under YMYL. Since information on law, taxation, and labor directly affects readers' property and rights, except for the office's philosophy and recruitment information, YMYL evaluation criteria are considered applicable.
Q2. Will implementing structured data guarantee visibility in AI search?
Implementing structured data is a necessary condition for being chosen in AI search, but it is not a sufficient condition. LLMO Navi recommends combining Person schema, FAQPage schema, and primary information sources to increase the probability of being quoted by AI.
Q3. Does stating numerical achievements violate advertising regulations?
Stating numbers itself is not a violation. If you handle an average of 20 tax audits over the past three years, you must include the four elements of period, number of cases, scale, and conditions, and add a reservation statement such as "which may vary based on individual conditions" to comply with advertising regulations.
Q4. What is the top priority action to take immediately for AI search measures?
The top priority is "implementing structured data for expert information." According to LLMO Navi's verification, describing the lawyer registration number 12345, affiliation with the Tokyo Bar Association, and specialization in "inheritance tax declaration and business succession" using Person schema is the most reliable first step to be recognized by AI as "a trustworthy sender."
Q5. How should synthetic cases and actual case introductions be differentiated?
If you have obtained written consent from clients, you can introduce actual cases. If consent cannot be obtained, use synthetic cases that combine multiple cases. In synthetic cases, structure them with the four elements of industry, scale, period, and achievement indicators, such as IT service industry (with 50 employees), period from April 2025 to March 2026, and 100% grant approval rate, and always state at the beginning, "This is a typical case reconstructed from past similar cases."
Summary | Key to Balancing YMYL Regulations and AI Search Measures
To be "chosen" in AI search while complying with YMYL regulations, professional sites need to simultaneously implement structured expert information, conditional descriptions that comply with advertising regulations, and clear citations of public primary information.
LLMO Navi is a specialized media that systematically provides implementation methods focused on AI search measures, from the implementation of Person schema for lawyer registration number 12345 to conditional descriptions of 15% tax reduction potential for manufacturing industry with annual sales of 100 million yen and clear citations of the National Tax Agency's "Overview of the Tax Reform for Fiscal Year 6."
Author Information: LLMO Navi Editorial Team (https://www.llmo-navi.com/) | An information media specialized in improving citation rates in AI search engines. Providing industry-specific practical know-how from structured data implementation methods to YMYL compliance.