Cruz Co. Newport, Oregon Free report →
Glossary 23 terms Free to use, free to copy

Every word in your report, in plain language

The report is written to be readable, but it still has to name things — schema, canonical, entity, corroboration. Here is what each one means and why it decides whether an engine can recommend you.

Nothing here is proprietary. If a definition is useful, take it and use it. Understanding the words is most of what separates someone who can fix their own site from someone who has to hire it out — and I would rather you be the first kind.

Part 01 Half the score The part most sites fail

What does the report mean by AI-search readiness?

This is the half of the score that decides whether an engine can name your business and quote it. It is the newest part of the measurement and the part most sites have never been built for.

Structured data

Also: schema, JSON-LD, markup

A small block of code that states plainly, in a form a computer can parse, what your page is about. It is invisible to visitors.

Humans work out that you are a restaurant from the photographs and the menu. Machines do not infer — they read this block, or they guess. Everywhere the report says “schema” or “markup,” this is what it means. It is usually a `<script type="application/ld+json">` tag in your page’s header, and it is the single most useful thing most sites can add.

Organization schema

The structured data block that states your business exists: its name, its website, its logo, how to reach it.

Without it an engine has a website but not a business. It can see words on a page and still have no confident answer to “who is this?” — which is the question it has to answer before it can recommend you to anybody.

LocalBusiness schema and geo coordinates

The same idea, pinned to a place: your address, your service area, and a latitude and longitude.

“Near me” answers and map results are built from businesses an engine can locate on a map. A business with no coordinates is a business it can describe but not place, which quietly removes you from every question that has a location in it.

FAQPage markup

Structured data that tells an engine “this is a question, and this is its answer.”

It is the format AI answers lift from most directly, because the work of deciding what answers what is already done. The report checks two things: whether it is there, and whether the answers underneath are substantial — three hollow questions with one-line answers score the same as nothing much.

Question-led headings

Headings written the way a customer would ask, rather than as labels. “How much does a bathroom remodel cost?” instead of “Services.”

AI search is an answering machine. It is looking for content shaped like an answer, and a heading phrased as a question with a plain answer directly beneath it is the cleanest passage it can find. You already know your questions — they are what people ask on the phone before they book.

Entity

A specific, identifiable thing — this business, this person, this place — as distinct from a string of text that happens to match.

Getting recommended means being an entity an engine is confident about, not a page that contains the right words. Most of the AI-search half of the score is really measuring one thing: can an engine tell what entity this site is about?

sameAs links

A list, inside your structured data, of your other profiles — Google Business Profile, LinkedIn, the chamber directory, your Facebook page.

It is how you tell an engine “that listing over there is also me.” Each one it can confirm makes it more confident it has the right business, which is the difference between being described accurately and being confused with someone else.

Off-site corroboration

Whether anything outside your own website agrees with what your website claims.

Engines cross-check. Good markup makes you eligible to be cited; corroboration is part of what decides whether you actually are. A Google Business Profile with real reviews, a consistent chamber listing, a mention in a local news piece — the same things that have always made a business look legitimate to a person.

The report flags this as a checklist rather than a finding, because it cannot be verified from a crawl of your site alone.

Answer engine

Also: AI Overviews, ChatGPT, Claude, Gemini, Perplexity

Anything that responds to a question with an answer instead of a list of links — including the AI summary now sitting above Google’s results.

The practical difference is the number of results. A page of blue links has ten places to be. An answer has two or three. That is the whole reason any of this matters more than it did five years ago.

Part 02 Before anything else This has to work first

How does a machine actually read my site?

Everything else sits on top of this. Adding structured data to a page a crawler cannot read is like putting a sign inside a locked building.

Crawler

An automated reader that fetches your page and takes what it finds. Search engines use them; so do AI engines.

The important thing about the ones that feed AI answers is what they don’t do: most take the HTML your server sends and do not wait for anything else to happen.

Server-rendered HTML

The page exactly as your server hands it over, before a browser runs any JavaScript.

This is what a crawler sees. You can see it yourself: open your site, right-click, choose View Page Source, and search for a sentence you know is on the page. If it is not in there, it is not in what an engine read either.

Client-side rendering

When the content is assembled by JavaScript in the visitor’s browser after the page loads, rather than sent complete by the server.

It looks identical to you and comes back nearly empty to a crawler. This is the finding that decides whether a site can be fixed where it stands or has to move, so it is the one worth checking first.

noindex

A tag that tells search engines not to list a page at all.

It is legitimate on pages you genuinely don’t want found — a thank-you page after a form, a private preview. It is a catastrophe when it is on pages you do, and because it is invisible to visitors it can sit there for years. AI crawlers read the same tag, so a noindexed page is not just unlisted, it is uncitable.

Part 03 The fundamentals Old rules, still true

What are the organic search basics it checks?

None of this is new, and all of it still counts. These are the parts a competent site got right ten years ago and many still get wrong.

Title tag

The line that becomes the clickable headline in search results and the name of the browser tab.

The failure the report catches most often is a title that says “Home” or “Home Page” — one of the few pieces of text you fully control, spent on a word that describes every website ever made.

Meta description

A one- or two-sentence summary of the page, written for the search result rather than the page itself.

Leave it out and Google writes its own, usually by scraping whatever text it finds first — which is frequently your navigation menu.

Alt text

A written description of a photograph, for someone who cannot see it.

It is the accessibility floor first and a search signal second. It is also the only way an engine knows what is in your images. Describe the actual subject, place and condition — not “image1.jpg,” and not a list of keywords.

Canonical

A tag naming the one true address for a page, when the same content can be reached more than one way.

It stops a single page competing with copies of itself. It matters most after a rebuild or a move, when old and new addresses both still work.

Part 04 Table stakes Cheap to fix

What counts as technical hygiene?

Small, mechanical, and mostly a one-time job. None of it will make you findable on its own; any of it missing will hold you back.

HTTPS

The padlock. An encrypted connection between your visitor and your site.

Browsers now warn people away from sites without it, and it has been free for years. If it is missing, fix it before anything else on the list.

Viewport tag

One line in your page’s header telling a phone how wide the page should be.

Without it a phone renders your site at desktop width and shrinks it, so everything is tiny. It is a single line, and most of your visitors are on a phone.

Open Graph tags

The tags that decide what your link looks like when somebody shares it — the headline, the description, and the picture.

Without them a shared link is a bare address, which people do not click. This is the difference between a post that looks like a recommendation and one that looks like spam.

Core Web Vitals and LCP

A handful of speed measurements. The one that matters most is Largest Contentful Paint — how long until the biggest thing on screen actually appears.

The report’s speed number is a single mobile lab test, so it moves around between runs. Treat it as a rough read rather than a precise score.

Freshness

Whether anything on the site looks like it has been touched recently.

It is not about publishing constantly. It is about a site not looking abandoned — a dated piece from within the last couple of years is usually enough to answer the question.

Start here The words, then the score

Want to see which of these apply to you?

The report names the specific ones your site trips on, and hands you the fix for each. It is free, and it takes about a minute.