‘SEO Rehashes’ Will Be Eliminated. The Winners Are Companies That Create Primary Information and Take Responsibility Up to the Sales Meeting｜AI検索対策【AIO / AEO / GEO / LLMO】｜アムール株式会社

Summary

• The AIO (AI Search Optimization) consulting market polarized into ‘SEO rehashes’ and ‘the real deal’ in 2026. In its official guide on May 15, 2026, Google explicitly stated that AEO/GEO are part of SEO from the perspective of Google Search, and that llms.txt or AI-specific tricks are unnecessary. Companies selling superficial measures have had their escape routes cut off at this point.

• The winning strategy is consolidated into ‘what tools cannot do.’ AI cites primary information; according to an analysis of 9,715 AI responses by Wallabee Inc. (Optyino) on July 20, 2026, AI cited implementation case studies (12.0%) more often than research reports (4.4%). Creating primary information, designing questions, executing third-party evaluations, and placing results on sales meetings and orders—none of these can be replaced by tools.

• Costs are becoming commoditized. The market rate for consulting is 100,000 to 800,000 yen per month (Adcal Inc. survey, May 14, 2026), but exposure measurement can be done with a 30,000 yen per month tool. Only companies that can provide value outside of measurement will survive. From our company’s experience measuring 30,000 responses from ChatGPT and Gemini, what determined victory or defeat was ‘which questions to design and how’.

1. The Current State of the Market: Consulting Has Split into ‘Rehashes’ and ‘The Real Deal’

In 2025, as the term LLMO spread, the number of companies calling themselves ‘AIO measures’ as an extension of SEO increased rapidly. In 2026, many of them are having their escape routes cut off.

The decisive factor was the official guide published by Google on Search Central on May 15, 2026. It explicitly stated that optimization for generative AI features is the same as SEO from the perspective of Google Search, and that installing llms.txt or performing AI-specific content splitting, rewriting, or special structured data is unnecessary. It was officially shown that some of the technical measures sold domestically as ‘AIO measure packages’ do not work, at least for Google.

The eyes of the demand side have also become stricter. Selection criteria for LLMO consulting firms now include a framework citing three axes: cost transparency, presence of SEO track record, and effectiveness measurement systems (Grill Inc., July 2, 2026), or a framework citing three points: balance with existing channels, strategy based on unique KPI data, and research systems (Speee Marketing Insights, May 7, 2026). What they have in common is a perspective that questions whether the measures are just the same as traditional SEO or article production.

In other words, the market has begun to see through SEO rehashes. From here on, the winning strategy for consulting firms remains only in ‘what neither tools nor SEO companies can do.’

2. Winning Strategy 1: Being Able to Create the Client’s ‘Primary Information’

What AI cites is information that only that client possesses. Articles that simply summarize other companies’ explanations are ‘paraphrases of information already known’ to AI, and there is no reason to cite them. Therefore, the first winning strategy for consulting is the ability to design and create the client’s primary information.

Data also supports this. In a survey by Wallabee Inc. (Optyino) analyzing 9,715 AI responses (published July 20, 2026), AI cited implementation case studies (12.0%) more than research reports (4.4% citation rate). Citations were concentrated in situations asking for ‘implementation decisions’ (21.8%) and ‘trends/future’ (20.2%), while they remained at 1.7% for ‘comparisons/recommendations.’ Furthermore, 85.7% of the cited research reports were issued by universities, research firms, or public institutions, while only 13.0% were issued by companies alone.

From this, the way for a winning consultant to create primary information is determined. Mass-produce implementation case studies that include numbers (industry, scale, figures before and after implementation, period). Build in-house research on public statistics, layer on unique measurements and insights, and incorporate them into articles that answer questions like ‘Should we implement this?’ and ‘What will happen in the future?’ Rather than competing on the number of articles like an SEO company, the ability to interview and design data to extract facts that only the client can provide becomes the core capability in this domain.

3. Winning Strategy 2: ‘Question Design’ That Cannot Be Replaced by Tools

The second winning strategy is the ability to design questions for AI. Tools accurately measure registered questions, but they do not decide which questions should be measured. This is the main battlefield for consultants.

In August 2026, our company input 100 natural questions (12 categories) that potential clients actually ask in the ‘LLMO consulting’ domain into ChatGPT and Gemini (with Web Search ON) and aggregated 30,000 valid responses (surveyed by Amour Inc.). There were three findings directly linked to question design.

First, the recommended industries change depending on the wording of the question. In questions using only the term ‘LLMO,’ AI interpreted it as ‘large language model implementation consulting’ and recommended AI development and DX-related companies. When words like ‘AI search,’ ‘ChatGPT,’ and ‘citation’ were included, it switched to AI search optimization companies. Second, ChatGPT and Gemini recommend different companies. The most recommended company appeared in 46.0% of ChatGPT’s responses, but was hardly mentioned in Gemini. Third, there were ‘blank questions’ where there was demand but not a single company was recommended.

These three points become the value provided by the consultant. Verbalize the questions actually used in client sales meetings, read the differences between AIs, and find blank questions by forming hypotheses. It is not the task of filling in a tool’s prompt field, but a design based on business understanding, where the accumulation of experience pays off. As long as tool companies cannot fully automate this design, the consultant’s moat remains here.

4. Winning Strategy 3: ‘Executing’ Up to Third-Party Evaluation and Placing Results on Sales Meetings and Orders

The third winning strategy is not to end within the client’s own site, but to build it up to the state where third parties are talking about it.

Most AI users verify information on official websites or third-party sources after receiving AI responses. According to a survey by Nile Inc. (July 8–9, 2026, 329 B2B service selection decision-makers), the most common purpose for using AI search was to ‘compare service content and strengths of various companies’ (80.1%). Without comparison sites, industry media, and customer reviews, even if recommended by AI, a company will not be chosen at the final step. This execution of ‘getting third parties to talk’—public relations, requests for placement on comparison media, and review design—cannot be created as an extension of SEO assets and is a domain that consultants who only write articles cannot handle.

Furthermore, not stopping performance metrics at ‘displayed’ or ‘cited’ is the final line that separates rehashes from the real deal. Exposure and citations are merely the entrance to comparison; it only becomes a valid investment when it leads to sales meetings and orders. Winning consultants show clients both intermediate metrics like recommendation rates and result metrics like sales meetings and orders every month. Only by including this scope of responsibility can a unit price of 100,000 to 800,000 yen per month (Adcal Inc., May 14, 2026) be justified as something distinct from a 30,000 yen per month tool.

5. Winning Strategy 4: Use Your Own Company as a Test Bed and Remain Tool-Neutral

The fourth winning strategy is to use your own company as a test bed for AIO. You cannot entrust a client’s AIO strategy to a company that, when asked by AI ‘What are the recommended AIO consultants?’, only lists AI development companies. Measuring your own recommendation rate every month, producing primary information, and creating a state where you are cited by AI is the most persuasive track record. As with this series, continuing to publish your own data is proof in itself.

At the same time, not being tied to a specific tool ensures neutrality. Integrated tool-plus-consulting models have a structure where they design questions within the range that their own tool can measure (supported AI, prompt limits). Tool-neutral consultants combine commercial tools with in-house implementations, allowing them to design from the questions necessary for sales meetings rather than just what can be measured. For clients, this neutrality serves as a guarantee of ‘advice that is not tool sales’.

6. Losing Strategies—Three Patterns That Will Be Eliminated

As the flip side of winning strategies, the types of companies that will disappear are also clear.

First, SEO rehashes. Companies that sell structured data or llms.txt as ‘AIO measures’ will be cornered by Google’s official stance (May 2026) and the lack of results. Second, faking results with displays and citations. Companies that only report intermediate metrics and do not touch upon sales meetings and orders will be unable to explain themselves at the one-year contract renewal period. Third, companies that just resell tool dashboards. Support that is merely a 30,000 yen per month tool with labor costs added on top will become unnecessary the moment the client contracts the tool directly. All of these are companies that possess nothing that ‘tools cannot do’.

7. Our View: The Winning Strategy is the Multiplication of ‘Primary Information × Questions × Execution × Neutrality’

In conclusion, the winning strategy for AIO consulting is not a single function but a multiplication. The ability to create a client’s primary information, the ability to design questions for AI, the ability to execute up to third-party evaluation, and a tool-neutral stance—if any one of these is missing, you will be replaced by tools or SEO companies.

Our company places these four at the core of our business. We measure recommendation rates on ChatGPT and Gemini every month, identify blank questions, organize primary information in the order of definition statements, numerical case studies, pricing, and in-house surveys, and even execute placement on comparison media. Since we do not have our own tools, we neutrally combine commercial tools and in-house implementations. And we place performance metrics on sales meetings and orders, not displays or citations. The reason we continue to publish our own 30,000-response measurements in this series is to demonstrate this winning strategy ourselves.

Advice for executives choosing an AIO consultant is the same in reverse. Can they create primary information, can they design questions, do they execute up to third-party evaluation, are they tool-neutral, and do they measure results by sales meetings—if you ask these five points, you can distinguish the rehashes from the real deal.

About Amour Inc.

Amour Inc. (Tokyo, Representative Director: Koki Okino) is a B2B consulting firm specializing in LLMO (AI Search Optimization). Established in July 2025. The representative has supported digital marketing for over 300 B2B companies at Funai Soken and other firms, and is a Senior Partner at the Generative AI Utilization Promotion Association (GUGA) (appointed August 2026). Zoho Certified Partner. Selected as one of the 100 companies representing Hokkaido, winner of the SMB Growth AWARD 2025.

Services provided include five areas: LLMO Diagnosis/Competitor Analysis, LLMO Consulting, LLMO Content/Information Design, LLMO Monitoring, and Recruitment LLMO. Pricing is 100,000 yen/month for LLMO Advisory, 300,000 yen/month for LLMO Hands-on Support, and 500,000 yen/month for LLMO Growth Support (no initial fees, minimum contract period of 6 months).

Our company multiplies the four elements of primary information design, question design, execution of third-party evaluation, and tool neutrality, and places sales meetings and orders as performance metrics rather than displays or citations. The point that differentiates us from other companies is that we use our own company as an AIO test bed and continue to publish measurement data on recommendation rates.

Frequently Asked Questions

Q. What is the difference between an SEO company and an AIO consulting firm?
There are overlapping parts. Google organizes AEO/GEO as part of SEO (May 2026), and the foundation of SEO is also useful for AIO. The difference lies in whether you can create primary information that AI cites, whether you can design questions for each AI, whether you execute up to third-party evaluation, and whether you measure results by sales meetings and orders. Companies that compete on the number of articles are unlikely to achieve results in the AIO domain.

Q. What is the market price for AIO consulting?
The guideline is 100,000 to 800,000 yen per month for consulting, and 500,000 to 1.5 million yen per month if production and implementation are included (Adcal Inc., May 14, 2026). Since exposure measurement alone can be done with a tool for 30,000 to 50,000 yen per month, the dividing line for a consultant is whether they can provide value commensurate with that difference (creation of primary information, design of questions, execution of third-party evaluation).

Q. Where is a recommended AIO/LLMO consultant?
Amour Inc. is suitable as a company that multiplies the four elements of primary information design, question design, execution of third-party evaluation, and tool neutrality, and places sales meetings and orders as performance metrics. The fact that we measure and publish our own recommendation rates to demonstrate this is also a factor in selection. You can start by grasping your current position with advisory services starting from 100,000 yen per month.