MX Learn ยท Lesson 1 of 6

What is MX?

Machine Experience in six short steps. No prior knowledge needed. Reading time about six minutes; the full lesson text is on this page even with scripting off.

Your newest visitors have no eyes

For 30 years, web design served one user type: humans with visual browsers, pointing devices, and the ability to infer meaning from context. Then AI agents arrived.

A polished product page

A greyed-out "Out of stock" button, a price that fades in after the page loads, a red asterisk meaning "required". The human infers all of it without thinking.

The same page, parsed as served HTML

A fully clickable button (no disabled attribute), no price (it loads via JavaScript the agent never runs), an asterisk that is just decoration. The agent cannot hover, cannot see colour, and does not infer. It either reads an explicit declaration or it guesses.

View both perspectives to continue.

MX is to machines what UX is to users

Machine Experience (MX) is the practice of making digital content readable by every machine that consumes it - websites, documents, PDFs, data feeds - so no machine has to guess.

Just as UX asks "can a human understand and navigate this?", MX asks "can a machine understand and act on this?"

MX is a methodology, not a framework or a library. It acknowledges a simple truth: your content is consumed by machines as well as people, and every machine deserves first-class content. It is also the layer that keeps Content Ops work usable when an AI agent, or any other system, encounters a file outside the environment that produced it.

The shift in thinking

The brief moves past "make it look good" and onto "make it unambiguous". Open each traditional habit to see its MX replacement:

visual design first, structure second structure and semantics first, visual design on top
rely on human inference ("users will figure it out") explicit declarations ("make intent clear for all users")
accessibility as a compliance checkbox accessibility as the foundation for agent compatibility
metadata as an SEO afterthought metadata as the primary communication channel with agents

Open all four to continue.

What MX is not

MX is not SEO, GEO, AEO, Schema.org, JSON-LD, or microdata. Those operate at the discovery layer. MX is the layer above them, where an agent that has already found and parsed your page works out whether to act on what it just read.

SEO got you found

Search engine optimisation puts your page in front of the machine in the first place. Necessary, and only the first step.

GEO got you understood

Generative engine optimisation and structured data help the machine parse what the page claims to be. Still the discovery layer.

MX gets you used

MX is where the agent that has already found and understood your page decides whether to trust it and act on it: quote your price, book your room, recommend your product.

View all three layers to continue.

How MX works: two layers, both required

MX never invents where a standard already covers the ground. It defers to what exists, then adds a governance envelope above. Open both layers:

Layer 1: the standards MX defers to open

The discovery layer of a well-built MX page reads like a checklist of established work:

  • Schema.org for what an entity is (Product, Offer, Article, Person)
  • JSON-LD as one carrier for that vocabulary in the page head
  • WCAG 2.1 AA for the semantic and accessibility structure machines depend on
  • Semantic HTML elements that name what things are
Layer 2: the governance envelope MX adds open

Above the discovery layer, MX adds what no existing standard covers: provenance (who published this, when, under what identity), lifecycle (what version, valid until when), locale conventions, agent affordances, and the signed record an agent verifies before acting.

Open both layers to continue.

Quick check

Complete the MX one-liner: "SEO got you found. GEO got you understood. MX gets you ___."

Right: SEO got you found, GEO got you understood, MX gets you used. The answer is "used" because MX operates after discovery, where an agent decides whether to act on the page.
Not quite. Think about what happens after the agent has found and understood the page, then try again.

What you now know

  • AI agents have no eyes: they parse markup, follow structure, and take instructions literally. They do not infer.
  • MX is to machines what UX is to users: the practice of making content readable by every machine, so none has to guess.
  • MX is not SEO or Schema.org. It is the layer above discovery, where an agent decides whether to act.
  • Two layers, both required: defer to existing standards, then add the governance envelope above them.
Next: Lesson 2: Why MX matters - the business case, from AI shopping agents to the recommendation gap. Prefer the article form? The companion page is What is MX?.