Tech Blog

A technical blog where we publish practical knowledge from building Web3, blockchain, and AI systems, written for IT engineers. We share first-hand accounts of the problems we ran into during implementation, design decisions, and verification results.

For a systematic guide to Web3, see our English Web3 Knowledge Hub; for talks, events, and other activity reports, see our Japanese blog. Most Tech Blog articles remain Japanese-only, while the English guides linked below are available in full.

Recent article themes

Our Tech Blog groups articles under themes including AIO (AI search visibility and optimization — how a site is found and cited by AI assistants and AI-generated search answers), SEO, and AI development. Recent themes include: how to select an AIO / AI-search vendor, the cost of AIO work, why strong technical capability often fails to come across on the web, checking whether AI crawlers are reaching a site via server logs, whether a JavaScript-rendered site's content reaches crawlers, comparing how much AIO work is possible on different CMS platforms, why AI search tends to recommend competitors instead of your company, why AI sometimes gets company information wrong and how to fix it, what an AIO diagnostic can and cannot tell you, measuring SEO/AIO results through to enquiries, an AI-search-engine optimization guide, AI writing and SEO quality, structured data (Schema/JSON-LD) implementation, and an AI Overviews guide.

Other recurring themes include GSC-driven content rewriting, whether to block AI crawlers such as OAI-SearchBot and GPTBot, E-E-A-T for B2B sites, how to identify search intent, local SEO/MEO basics, AI-driven local discovery, review and rating strategy, e-commerce product-page SEO/AIO, growing branded search, mobile UX and Core Web Vitals, and how to choose an SEO or web-production partner. More recent additions cover what drives the cost of B2B website improvement and how a 90-day programme is sequenced, how to write an improvement specification that a production agency or in-house engineer can actually implement, and how to share GA4 and Search Console access with an outside firm under least-privilege settings. We also cover investigating Search Console's "crawled - currently not indexed" state, notifying Bing of updates with IndexNow, the SEO side effect of Vercel Skew Protection's ?dpl= parameter, a 20-item self-diagnosis checklist for B2B sites, why a manufacturer's technical content is rarely cited by AI search, information design for professional-services firms, the difference between being cited and being absorbed into an AI answer, why AI visibility cannot be measured in a single run, and why optimising for generative AI search is an extension of SEO rather than a separate discipline.

On the AI-development side, we have published write-ups on connecting business tools to AI agents via MCP (Model Context Protocol), keeping structured data intact across Next.js and Strapi, tracking AI citations, running RAG (retrieval-augmented generation) in production, running our own AI-agent operations ("Netsujo's operating OS"), building an in-house SEO analytics dashboard that combines GA4 and Search Console, our record of spec-driven development and parallel project work with Claude Code, and a Japanese-only account of how Netsujo SIGNAL grew out of improving our own two websites with AI agents rather than being planned as a product to sell. Japanese-only pieces also cover how an AI adoption assessment is run, fixing Japanese headings that wrap badly on mobile, why a company that is never discovered in search or generative AI is never compared in the first place, why AI agents remove the wait for implementation but concentrate the harder calls — scope, priority, when to ship, when to stop — on one person, why splitting design, implementation and audit across ChatGPT, Claude Code and Codex helps but should not be applied to every change in series, and what we can take on for AI development and for system development from our base in Kyoto.

Seven long-form guides are also available in English: diagnosing why a B2B website is not generating enquiries, converting SEO reports into implemented improvements, identifying why AI search does not recommend a company, what determines the cost of B2B website improvement over a 90-day roadmap, how to write an improvement specification for the web, how to share GA4 and Search Console access safely, and why SEO agencies ask us to take on the coding layer of their improvement plans.

Diagnose why a B2B website is not generating enquiries

Turn SEO reports into implemented improvements

Identify why AI search does not recommend a company

Understand B2B website improvement cost and the 90-day roadmap

Write an improvement specification that can be estimated and implemented

Share GA4 and Search Console access safely with an outside firm

See why SEO agencies ask us to take on the coding layer

The Tech Blog's full article archive is currently Japanese-only.