SHANNON合同会社
日本語 Contact

Discovered and cited
correctly by AI.

Generative AI such as ChatGPT, Perplexity, and Google AI Overviews increasingly researches companies and services to answer questions. AI search optimization (AIO / GEO) sets up everything from AI-crawler access, to an llms.txt that summarizes your site structure for AI, to structured data and server-side crawlability — so AI discovers and cites your business correctly.

Free consultationDownload materials
Pain points

Do any of these sound familiar?

Whether AI search doesn’t surface your company, or describes it incorrectly — that’s the right time to talk.

When you look yourself up in ChatGPT or Perplexity, you don’t show up, or you’re described with outdated information.
AI summarizes or cites your business and services incorrectly.
You’ve done SEO, but you don’t know what to do for AI search.
You’re not sure whether your robots.txt and structured data are set up for AI.
You’re worried your server’s WAF or access restrictions might be blocking AI crawlers.
Reasons

How we measure results

Traffic via AI can’t be captured in a single number the way a search ranking can. That’s exactly why we combine several perspectives to track whether it’s growing and whether you’re being cited correctly. We’ve seen it ourselves — as we kept improving our own visibility, traffic from AI search actually grew.

01

How we measure AI-driven traffic

Traffic from links inside answers in ChatGPT, Perplexity, and the like records the AI service’s domain as the referrer. We use that as a signal to isolate and understand traffic that appears to come via AI.

02

How to read it in GA4

We segment AI service domains by source / medium and track sessions, entry pages, and what users do next. While volumes are small we read it as a trend; as it grows, we monitor it as a fixed measure.

03

Tracking how you’re cited, over time

Beyond traffic, we regularly search the major AI tools to check how you’re actually described and cited. The accuracy of those descriptions often improves before the numbers move.

Scope

What we cover

From an AI-visibility assessment, to crawl infrastructure, structured data, and server-side crawlability, through to continuous measurement.

01 Assessment

AI-visibility assessment

We run real searches on the major generative AI tools and measure whether you show up at all, how you’re described and cited, and whether the information is outdated. We also surface issues in robots.txt, structured data, and server settings, and prioritize improvements against cost and impact. We start by making the current state visible in numbers and in the AI’s actual answers.

  • Real measurement of presence and citations across major AI
  • Surfacing incorrect and outdated information
  • Review of robots.txt and structured data
  • Checking for server-configuration issues
  • Improvement-priority map
02 Crawlability

AI-crawler access

We allow the major AI crawlers — GPTBot, OAI-SearchBot, PerplexityBot, Google-Extended, and others — in robots.txt and declare the sitemap. We also prepare an llms.txt that summarizes your site structure and key points for AI, so it can grasp the whole site correctly in a short time. We serve an llms.txt on our own site, too.

  • Allow settings for major AI crawlers
  • Declaring the sitemap
  • Creating and serving llms.txt
  • Controlling areas you don’t want crawled
03 Structured data

Structured-data setup

We set up JSON-LD such as Organization, FAQ, Article, and BreadcrumbList so AI can interpret page content without misunderstanding. By expressing operator information and the relationship between your business and services in a machine-readable form, we make it harder for AI to mistake which company and which service is which.

  • Making Organization / operator information explicit
  • FAQ, Article, BreadcrumbList
  • Clarifying the business-to-service relationship
  • Validating the markup
04 Server

Checking server-side crawlability

We check whether WAF, access restrictions, overseas-IP blocking, or bot measures are turning AI crawlers away, and make sure crawling actually gets through. Even if you allow it in robots.txt, it won’t reach AI if the server blocks it. We verify — via logs and real requests — that crawling truly gets through.

  • Checking crawlability against WAF and bot measures
  • Reviewing overseas-IP and access restrictions
  • Confirming crawl activity in access logs
  • Handling excessive crawl load
05 Measurement

Measurement & continuous improvement

We track, as a fixed measure, how the major generative AI tools describe and cite your company and services, and alongside that we follow changes in traffic that appears to come via AI in GA4. Because we can’t control how the AI behaves, we can’t guarantee you’ll show up — but by continuing to take steps that raise visibility, we can genuinely create growth in traffic from AI search.

  • Fixed-point measurement of AI descriptions and citations
  • Tracking AI-driven traffic in GA4
  • Continuing to adjust as things change
  • Sharing regular reports
Process

How we work

We measure the current state first, set priorities, and then do the setup. We work fully remote in principle.

  • 01Interview & baseline measurement — we measure how you actually appear across the major generative AI tools, and learn the search contexts you want to target.
  • 02Assessment report — we surface issues in robots.txt, structured data, and server settings, and present them in priority order.
  • 03Implementation — we set up crawl infrastructure (robots.txt / Sitemap / llms.txt), structured data, and server-side crawlability.
  • 04Measurement & continuous improvement — we track how AI cites you as a fixed measure and keep improving.
Plans

Engagement formats

Rather than setting up everything at once, we measure the current state, set priorities, do the setup, and then keep tracking it over time. We design the scope to fit each engagement.

Diagnose

Start by assessing AI visibility

  • Real measurement of presence and citations across major generative AI
  • Surfacing incorrect and outdated information
  • Review of robots.txt, structured data, and server settings
  • Presenting an improvement-priority map
Build

Crawl infrastructure & structured-data setup

  • Allowing major AI crawlers and declaring the sitemap
  • Creating and serving llms.txt
  • Structured data such as Organization / FAQ / Article
  • Checking crawlability against WAF and access restrictions
Measure

Fixed-point measurement & continuous improvement

  • Fixed-point measurement of descriptions and citations across major AI
  • Tracking traffic that appears to come via AI in GA4
  • Continuing to adjust as things change
  • Sharing regular reports
Comparison

How it differs from traditional SEO

AIO / GEO doesn’t replace SEO — it sits alongside it. Much of the underlying work overlaps, but the goal you’re aiming for and how you evaluate it are different. We set things up on the premise of doing both.

Traditional SEO

Rank high in search results

  • The goal is ranking on the search results page
  • Users click a link and visit
  • Evaluated by rank, click-through rate, and traffic
  • Optimized around keywords
AIO / GEO

Be cited correctly in AI answers

  • The goal is appearing and being cited within AI answers
  • AI summarizes it and uses it directly in its answer
  • Evaluated by presence, accuracy of description, and AI-driven traffic
  • Centered on crawlability and an unambiguous structure
FAQ

Frequently asked questions

Where SEO aims to raise your ranking in search results, AIO / GEO aims to have you appear and be cited correctly within the answers of generative AI. It centers on building the foundation that lets AI crawl and retrieve your site, and on structuring content so it’s interpreted without misunderstanding — and it works alongside your SEO efforts.
We don’t guarantee that. Because we can’t control how the AI behaves, we can’t promise you’ll show up reliably. What we can do is create a state that’s easy to discover and retrieve and hard to misinterpret, and keep taking steps that raise visibility. We tell you that honestly up front and proceed from there.
Yes. robots.txt, Sitemap, structured data, llms.txt, and server settings don’t depend on how the site is built, so we can handle non-WordPress sites too. We first learn about your current setup, then propose how to proceed.
It varies with the site’s size, current state, and scope, so we handle it on a case-by-case basis. You’re welcome to start with an assessment of your current AI visibility. Feel free to get in touch.
Related

Related services

AI

AI Agent Development

To embed generative AI into your operations or product, we cover PoC through production and connect safely to your existing systems via MCP.

See AI agent development Analytics

Analytics & Measurement Audit

To track AI-driven traffic correctly, start with the measurement foundation. We review your GA4 setup, tags, and data reliability with a third-party eye.

See analytics & measurement audit Team

Engineering Agency

If you want to hand off development beyond site improvements, we run alongside you as one of your in-house members, from design and implementation into operations.

See engineering agency
Get in touch
Contact

Let’s talk — the first consultation is free.

Even if your requirements aren’t fixed yet, that’s fine. We reply within 2 business days.

You can also reach us by phone (050-1794-9651, automated voice; we call back on business days). For detailed enquiries with budget and timing, please use the contact form. Note: we do not accept sales solicitations.