The invisible users are here
AI agents don’t click, don’t scroll, and don’t forgive ambiguity. They parse your HTML, evaluate your metadata, and make decisions in milliseconds. If your content isn’t structured for them, they move on.
MX Books — Machine Experience by Tom Cranstoun
Making the web work for AI agents and everyone else.
The web is no longer consumed only by people. AI agents are becoming a new consumption layer for enterprise digital platforms — interpreting, transforming, comparing and acting on information on behalf of users, teams and systems.
That changes what “quality” means. Not just user experience.
AI agents are visiting your site right now.
When they succeed, they build computational trust and return — recommending you more confidently each time. When they fail, they route around you permanently — no analytics warning, no second chance, no angry email explaining what went wrong.
Now you can see them
Machine Experience makes the invisible visible. Understand how agents interpret your pages, where they succeed, and where they fail — before your competitors do.
Now they can see you
MX gives you the language, principles, and patterns to design digital systems that machines can read, trust, and act on — reliably, safely, and consistently.
The MX Book Series
Three books. One discipline. From introduction to full specification.
MX: The Intro
The starter guide to building a web that works for AI agents and everyone else.
Free — Advance Preview
ISBN: 978-1-067638-41-2
Download Free
MX: The Handbook
Everything you need to design the web for AI agents — and everyone else.
£35.00
ISBN: 978-1-067638-40-5
Buy Now
MX: The Protocols
The formal patterns and field-level specifications for an AI-readable web.
Available 1 July 2026
What You Will Learn
The MX Framework
How to structure HTML, metadata, and Schema.org so every AI agent can read your content without executing JavaScript. The served HTML is the product.
The Five-Stage Journey
Discovery, citation, comparison, pricing, and purchase confidence. Each stage has explicit patterns that determine whether an AI agent recommends you or routes around you.
Governance & Compliance
Content lifecycle metadata, AI agent permissions, data sovereignty, and the MX governance layer that sits on top of established standards.
Implementation Patterns
Practical code patterns for semantic HTML, Schema.org JSON-LD, accessibility, llms.txt, robots.txt, and the full metadata stack.
Enterprise Platforms
How to apply MX to content management systems, e-commerce platforms, and digital experience platforms.
Measuring Success
How to audit your site against MX principles, benchmark against competitors, and measure the business impact of machine-readable content.
Machine Experience (MX)
MX is the discipline of designing digital systems so that machines can read, trust, and act — reliably, safely and consistently — across channels, devices and emerging agent platforms.
The invisible users are here. Now you can see them. Now you know how to make sure they can see you.
Get the Books
The Author
