Free Web Diagnosis Report
We publish what the free diagnosis actually returns, using our own site, netsujo.jp, as the example. No fictional company and no invented score.
This report is not a copy of an external AI's answer. It is produced by the Netsujo SIGNAL diagnosis agent, which reviews the official site's technical structure, key pages, business and trust information, enquiry paths, and how the site appears in AI search, and organises the confirmed facts and priorities.
- Company
- Netsujo Inc.
- Diagnosis target
- https://netsujo.jp
- Diagnosis date
- June 30, 2026
- Pages reviewed
- 7 pages
- Method
- Netsujo SIGNAL's specialised AI diagnosis agent — public-information crawl, key-page analysis, AI-search observation, and evidence verification.
01 / Overall assessment
Netsujo Inc., an implementation-led business-development company in Web3 and AI, stands out for measures that reduce a BtoB buyer's pre-commitment anxiety: fixed-price packages, explicit fit/no-fit guidance, and FAQ coverage. The technical foundation is also in place, with JSON-LD structured data, a sitemap, robots.txt, and llms.txt all implemented, showing attention to both search and AI. On the other hand, published case studies number only 3 (one still in progress), which limits how strongly the site conveys business scale and trust; the contact page also exists at several near-duplicate URLs, which can dilute how a page is evaluated.
02 / Areas assessed
| Area | Score | Comment |
|---|---|---|
| Findability in search | 62/100 | The top page's title and meta description are specific, but individual service pages vary in how well their meta descriptions are written. |
| Readiness for AI search | 75/100 | llms.txt, JSON-LD, and a sitemap are all in place, giving a structural foundation; content coverage is the next lever. |
| How it reads to a prospective customer | 70/100 | Fit/no-fit guidance, FAQs, and staged consultation entry points are handled carefully, but the small number of case studies limits the overall weight of trust signals. |
| Enquiry path | 65/100 | Multiple CTAs and a free-consultation path are clear, but the contact page exists at several duplicate URLs, which leaves room for tidying. |
| Structure and machine readability | 72/100 | JSON-LD, sitemap, robots.txt, and llms.txt are all implemented, so the foundation is solid. |
Scores are a simplified assessment based on public information.
03 / What is working well
- Fixed-price packages that remove pre-commitment uncertainty
- Confirmed: the top page and meta description state fixed packages with amount, duration, and deliverables — Web3 PoC design at JPY 800,000, a smart-contract pre-procurement audit at JPY 400,000, and an AI-applicability assessment at JPY 300,000. Why it matters: this directly addresses cost uncertainty, the biggest early-stage drop-off factor in BtoB, and works as a point of differentiation.
- Clear fit/no-fit guidance builds trust
- Confirmed: the top page lists "cases we are a good fit for" and "cases we are not" side by side, explicitly excluding customers who want the cheapest possible order against an already-fixed spec. Why it matters: being selective about who the company works with tends to support trust and reduce mismatched sales conversations in a BtoB context, and reads as evidence of specialisation.
- Technical AI readiness — llms.txt, JSON-LD, and a sitemap
- Confirmed: llms.txt, JSON-LD, sitemap.xml, and robots.txt are all in place. Why it matters: generative AI and search engines have a foundation for understanding the site's structure and content — a base to build on as AI-driven traffic grows.
04 / Top-priority issues
Listed in priority order, together with the effect we would expect from addressing each one.
- The number of public case studies is small, limiting the depth of trust signals
- Fact: only 3 case studies are published in detail (one still in progress), plus some project cards under "other engagements" marked as pages to be published later. Why it matters: in BtoB decision-making, case studies are a primary basis for trust, but with this few, a visiting prospect cannot easily find a comparable example or judge the range of industries and company sizes served. Direction: prioritise turning the "to be published" engagements into full pages, and build up the count and diversity by industry, phase, technology, and budget range until filtering actually becomes useful — even anonymised or summary-only listings can help show volume and variety. Expected effect: as more examples close to a visitor's own industry or size appear, visitors should find it easier to judge similarity to their own situation. (Reference: https://netsujo.jp/works)
- The contact page exists as duplicate content across several URLs
- Fact: https://netsujo.jp/contact, .../contact?type=development, and .../contact?type=government were all found to have the same title, H1, meta description, and body content. Why it matters: if the query-parameter URLs get indexed individually, it can create duplicate-content dilution, and because no content is tailored per referral type, it may add friction after a visitor reaches the contact form. Direction: point canonical tags on the query-parameter URLs to the base URL (/contact), and consider showing different content per type — for example, sample enquiries for government clients versus development clients. Expected effect: search engines should be able to evaluate the representative URL more clearly, and tailoring content by visitor type could lower the psychological barrier to enquiry. (Reference: https://netsujo.jp/contact)
- Service-page meta descriptions do not clearly convey what differentiates each service
- Fact: the PoC-support page's meta description reads like a list of keywords ("for those considering PoC support, a PoC development company, new-business PoC design"), whereas the top page's meta description states specific value (fixed pricing, end-to-end support); the contact page's meta description is generic ("enquiries about Web3 consulting, systems development, offshore development, and PoC support"). Why it matters: whether someone clicks through from a search result depends heavily on the meta description, and if a page cannot convey a differentiated value versus competitors, click-through is unlikely to improve. Direction: rewrite each service page's meta description around one or two of Netsujo's actual strengths (fixed pricing, defined exit criteria, support from the concept stage) rather than generic keywords. Expected effect: how the page's content comes across in search results should improve, moving closer to language that resonates with the target audience. (Reference: https://netsujo.jp/services/poc-support)
05 / What to do next
Now:
- Set canonical tags on the query-parameter contact URLs (/contact?type=xxx) to point at /contact, to consolidate duplicate content.
- Turn the "to be published soon" case studies on the case-studies page into full pages, moving toward a count that shows range across industries and phases.
- Rewrite the meta descriptions for the PoC-support page, the contact page, and others around Netsujo's actual strengths, rather than generic keywords.
Next:
- Once the case-studies filters (by industry, budget range, phase, and technology) have enough entries to be useful, build landing pages for specific search intents, such as "Web3 PoC support" or "local-government DX consultation."
- Restructure the contact page so it can branch by enquiry type, showing relevant case studies for government clients and development clients respectively.
- Review how often the technical blog and knowledge pages are updated, and put a regular publishing rhythm in place to build up a body of content.
Long-term:
- As engagement case studies accumulate, split out industry-specific service pages (finance, healthcare, local government, and so on) with tailored messaging for each audience.
- Extend the range of JSON-LD schema types (Service, FAQPage, BreadcrumbList, and others) to match each page's content, improving the completeness of structured data.
- Turn community activity (Connpass events, participation in Chain Up KYOTO, and similar) into knowledge articles and case content on a regular basis, to make specialisation visible over time.
What is not included in the free diagnosis
- Analysis of real data using GA4 and Search Console
- Finished page copy
- Structured data
- Measurement specification
- Production implementation
- This free diagnosis is limited to crawling public information (the top page plus up to 6 key pages) and an AI's first-pass findings.
- It does not guarantee outcomes such as search ranking, inclusion in generative-AI answers, enquiry volume, or sales.
- We do not provide an overall score (no single "XX/100" number). The per-area scores are indicative only.
- It does not include finished implementation code or complete finished copy (direction for improvement only).
- It does not include non-public data such as GA4 or Search Console, competitor comparisons, or a detailed technical audit — these are part of the paid support.
If you want to take the next step
For deeper root-cause analysis, implementation code, page copy, and measurement design, Netsujo SIGNAL's paid support covers:
- Analysis of real GA4 and Search Console data
- Draft improvements to titles and meta descriptions
- CTA and enquiry-path design
- Conversion measurement design
Netsujo Inc. / Netsujo SIGNAL. This report is a simplified diagnosis based on publicly available information. It does not guarantee search ranking, inclusion in generative-AI answers, enquiry volume, or any other outcome.