Agent-ready web
How to write an llms.txt that AI assistants actually use
Most llms.txt files are sitemaps in disguise. How to write one an AI assistant can answer from: summary, facts, links, Markdown pages, testing and upkeep.
Guides for websites that work with AI
Practical, sourced articles on AI agents for websites, the agent-ready web, AI sales and support, and AI search — written for website owners and marketers.
All articles
Agent-ready web
How to write content AI assistants quote accurately
Why AI assistants misquote websites and how to write pages they quote correctly: self-contained passages, complete facts, one source of truth and plain HTML.
Guides
Interactive quote calculators: why they beat contact forms
Why B2B buyers prefer an instant estimate to a contact form, how to design a quote calculator that qualifies leads, and how to test it against your form.
More articles
AI support
7 min readKeeping a website AI assistant accurate as content changes
Prices, policies and products change, and your AI assistant must keep up. How re-syncing works, what goes stale first and a routine that keeps answers right.
AI support
7 min readKnowledge gaps: using unanswered questions to fix your site
Unanswered questions from chat and site search show exactly what your website is missing. How to collect them, sort them into gap types and fix each one.
AI sales
7 min readLead scoring for website conversations: a simple model
A transparent lead scoring model for website chat: fit and intent points, negative signals, thresholds that trigger action, decay and how to calibrate it.
Agent-ready web
8 min readllms.txt explained: who reads it and how to write one
llms.txt is a Markdown map of your site for AI agents. What the v2 proposal says, who reads the file as of October 2026, and a template you can copy today.
Agent-ready web
9 min readllms.txt, robots.txt and sitemaps: what each file tells AI
robots.txt sets permissions, sitemap.xml lists pages and dates, llms.txt offers a reading list. What each file tells AI systems and how to keep them in sync.
Guides
8 min readMeasuring an AI agent on your website: metrics that matter
Which numbers show whether a website AI agent works: usage, answer quality, outcomes and impact against a holdout, plus the vanity metrics to stop reporting.
AI sales
7 min readQualify inbound leads with AI without annoying visitors
How to qualify website leads with an AI agent without pop-ups, interrogations or gated answers: what to ask, when to ask it, and how to route each visitor.
Agent-ready web
7 min readrobots.txt for AI bots: which to allow and which to block
Which AI crawlers to allow or block in robots.txt, sorted by what they do: training, AI search or user requests. Copy-ready files and the mistakes to avoid.
AI search
7 min readSite search analytics: what zero-result searches tell you
Zero-result searches show what visitors want and your site does not deliver. How to track them in GA4, read the patterns and fix search, content and pages.
Guides
7 min readThe cost of an AI agent on your website, explained
What an AI agent on your website really costs: model usage per answer, vendor pricing units, your team's time and extras, with a worked example and checklist.
AI support
8 min readTrain an AI chatbot on your website and keep it accurate
What training a chatbot on your site really means, how to prepare, chunk and test your pages, and the routine that keeps its answers accurate after launch.
Guides
7 min readTurning visitor questions into content ideas for SEO
How to mine website chat logs for content ideas: export and anonymize visitor questions, cluster them, check search demand and turn the gaps into useful pages.
Guides
8 min readVoice agents on websites: use cases, costs and pitfalls
When a talking website helps and when it gets in the way: realistic use cases, per-minute costs as of October 2026, and the pitfalls to plan for at launch.
Agent-ready web
7 min readWebMCP tools for B2B sites: which actions to expose first
Which actions a B2B website should offer AI agents through WebMCP first, which to hold back, and how to name, describe and safeguard each tool, with examples.
AI support
8 min readWebsite chatbot vs AI agent: what is the difference?
A chatbot replies; an AI agent decides on steps and takes actions. A plain guide to the difference on websites, with examples, risks and a checklist.
Agent-ready web
8 min readWhat is WebMCP? A plain-English guide for website owners
WebMCP is a draft browser API that lets your site hand AI agents named tools instead of making them guess at buttons. What it is, its status and what to do now.