Technical SEO and AI search visibility
Sites that search engines and AI assistants can crawl, understand and cite, with the checks built into every deploy.
Engineering for search results and AI answers: server-rendered, fast pages; clean URLs and one-hop redirects; structured data that describes your business, services and people consistently; sitemaps with honest dates; llms.txt, a facts page and, where it fits, an MCP server so AI assistants can read you directly. This site is the worked example, checked on every build.
- 01
A technical audit: rendering, indexing, structured data, speed and internal links
- 02
Fixes in your code, with CI checks so they stay fixed
- 03
Structured data for your organization, services, people and content
- 04
Deliberate rules for search and AI crawlers, a facts page and, last, llms.txt
- 05
A monthly measurement plan in Search Console and Bing Webmaster Tools
Who this is for
- Products whose pages rank poorly because crawlers see little HTML
- Businesses that AI assistants describe wrongly, or not at all
- Sites planning a redesign or a domain move without losing traffic
Who this is not for
- Link buying or mass-produced content
- Promises of a ranking position
What this service is
Technical SEO and AI search visibility is the engineering that lets search engines and AI assistants crawl a site, understand it and cite it: server-rendered, fast pages; clean URLs and one-hop redirects; structured data that describes your business, services and people consistently; sitemaps with honest dates; and deliberate rules for AI crawlers. It is for products whose pages rank poorly because crawlers see little HTML, businesses that AI assistants describe wrongly or not at all, and sites planning a redesign or a domain move.
This site is the worked example, checked on every build: titles, descriptions, canonical URLs, structured data and links are tested in CI.
Scope
The audit comes first and is fixed-price. I crawl the site the way a search engine and an AI crawler would, page type by page type, and record what they actually receive: status codes, rendered HTML, titles, canonicals, structured data, internal links and speed. The fixes follow in stages, quoted after the audit, in the order that unblocks the most.
What is measured and built
The foundations are the same for search and AI answers. Google says the best practices for SEO remain relevant for AI Overviews and AI Mode, and that there are no additional requirements or special optimizations needed to appear in them: a page must be indexed and eligible to show with a snippet. So most of the work is crawlable HTML, fast pages, accurate structured data and clear, specific content.
For AI assistants beyond Google, crawler rules matter. OpenAI, for example, states that sites opted out of its OAI-SearchBot crawler won't be shown in ChatGPT search answers, and that this is a separate setting from GPTBot, which concerns training. I set those rules deliberately with you, and make sure your firewall isn't blocking the crawlers you want. Files such as llms.txt cost little and serve some coding tools, but Google says you can ignore them for Google Search, so they come last.
Checks run in CI so fixes stay fixed: unique titles and descriptions, one H1, valid structured data, no broken internal links, and the redirect list. Measurement uses Search Console and Bing Webmaster Tools by country, plus a fixed set of questions asked of the main assistants each month.
What it is not
It is not link buying, mass-produced content or a promise of a ranking position; no one can honestly make that promise. It is not content writing either: the expertise in the content has to be yours, though I can help shape it.
How the engagement runs
Crawl and measure
What search engines and AI crawlers actually receive, page type by page type.
Fix what blocks indexing
Rendering, canonicals, redirects, sitemaps and speed, in the order that matters.
Describe the business
Consistent facts and structured data about who you are, what you offer and where.
Measure monthly
Queries and clicks by country, and a fixed set of questions asked of the main AI assistants.
Technologies I use for this
- Next.js
- Schema.org JSON-LD
- Google Search Console
- Bing Webmaster Tools
- IndexNow
- Lighthouse CI
- llms.txt
- Model Context Protocol
How this works in your market
How this works in the United States
For US teams, Google and Bing are the engines to measure, and ChatGPT search, Perplexity and Copilot the assistants to check. Search Console and Bing Webmaster Tools are verified and sitemaps submitted in the first week of the engagement, so there is a baseline to measure against.
How this works in the United Kingdom
For UK teams planning a redesign or a domain move, the work starts with the full list of URLs that rank and convert today, and ends with every one of them redirecting in one hop to its new home, tested before launch. Google's documentation on site moves is explicit that permanent redirects don't lose PageRank; the risk is in the ones you forget.
Questions about Technical SEO and AI search visibility
What is AI search visibility?
Whether assistants such as ChatGPT, Claude, Perplexity and Google's AI answers find, understand and cite your site. It rests on the same foundations as search (crawlable HTML, clear facts, consistent structured data) plus a few AI-specific files.
Do you write the content too?
I build the technical foundation: templates, structured data, internal links and checks. The expertise in the content has to be yours, and I can help shape it.
Can you guarantee rankings?
No one honestly can. I fix what is measurably wrong and set up the measurements, so you can see what changed and why.
Do we need llms.txt?
It's cheap to add and some coding tools read it, but Google says you can ignore it for Google Search. It comes after the work that does matter: crawlable pages, accurate facts and structured data.
Should we block AI crawlers?
That's a business decision, and it can be made separately for search and for training. Blocking a search crawler such as OAI-SearchBot keeps you out of that assistant's answers; blocking a training crawler doesn't.
Will structured data get us rich results?
Some, such as breadcrumbs. Google no longer shows FAQ rich results, so FAQ markup now mainly helps other engines and AI extraction. The aim is accuracy, not decoration.
Case studies behind this service
- Case study: TopGear India CMS · CMS migration and modernization
- page load time, before → after
- 1–3 min → 3 s
- Case study: InfluencerX · Real-time voting platform
- users over the campaign
- 1.6M+
Related writing
- Article: Is llms.txt worth it in 2026? What Google, OpenAI, Anthropic and Perplexity document · 11 October 2026
What the llms.txt proposal is, what each company’s own documentation says about it, what Ahrefs’ crawl of 137,210 domains found, and what to do instead.
- Article: A glossary for buyers of engineering work: AI, performance, security and data protection · 10 October 2026
Fifteen terms that come up when you hire for AI features, performance work, security audits or cross-border projects, each defined in two sentences, with why it matters.
Where this work happens
- Working with teams in United States · New York · Chicago · San Francisco
Founders and CTOs who need senior engineering without a full-time hire: MVPs, AI features, performance work and rescuing apps built in a hurry.
- Working with teams in United Kingdom · London
Scale-ups and agencies that want a senior engineer for platforms, performance and security, with most of the working morning in common.