Page title — agentsite

# Page title

The <title> tag in the page head. Highest-weight head-tag signal for "what is this page" — generic or boilerplate titles lose every comparison against pages that name their topic.

By AgentSite · 2 min read · Updated 2026-05-24

A page title is the text inside the `<title>` element in the page `<head>`, usually mirrored in `<meta property="og:title">` for Open Graph consumers. It is the highest-weight head-tag signal an answer engine has for "what is this page." A unique, topic-naming title raises every other ranking signal; a generic title ("Welcome," "Home," the site name alone) drops them all.

## The two tags

```html
<title>What is FAQ schema — AgentSite</title>
<meta property="og:title" content="What is FAQ schema — AgentSite">
```

[MDN](https://developer.mozilla.org/en-US/docs/Web/HTML/Element/title) defines `<title>` as "the document's title that is shown in a browser's title bar or a page's tab" and notes that "titles should be unique to every page of a website, ideally surfacing the primary purpose of the page first, followed by the name of the website." Search engines display about the first 55–60 characters.

[The Open Graph protocol](https://ogp.me/) treats `og:title` as one of the four required tags (alongside `og:type`, `og:image`, `og:url`). Social and AI consumers prefer the OG variant when present.

## What Google actually does

[Google's title-link documentation](https://developers.google.com/search/docs/appearance/title-link) says the engine "uses a number of different sources to automatically determine the title link," including the `<title>` element, on-page H1, prominent styled content, and `og:title`. The `<title>` is the first signal in the stack but not the only one. Google's recommendation: "Make sure every page on your site has a title specified in the `<title>` element… Don't engage in keyword stuffing; there's no reason to have the same words or phrases appear multiple times."

For AI retrieval the same logic holds. The page title is the engine's fastest read for the page's topic before it parses anything else; mismatch between the title and the body weakens both.

## Anti-patterns

Three failures show up reliably:

1.  **Site-name-only titles.** Every page reads `<title>AgentSite</title>`. The engine has no per-page distinguisher and falls back to extracting from the body — which on a SPA is often empty.
2.  **Boilerplate templating.** `<title>{{ page.title }} | AgentSite</title>` where `page.title` is missing on most routes, producing `<title>| AgentSite</title>` (literal pipe-and-name). Mechanical fix; nobody notices until an engine flags it.
3.  **Keyword-stuffed titles.** `<title>AEO Tool, AEO Platform, AEO Service, AEO Software — AgentSite</title>`. Google's documentation calls this out directly; AI extractors discount the page accordingly.

The honest pattern: one unique title per page, topic first, site name last, kept inside ~60 characters.

## Where this fits

Page title is a Layer 3 head-tag signal that doubles as a Layer 4 content-quality signal — the engine reads it as both metadata and a content summary. It pairs with [meta description](/meta-description) and [og:image](/og-image) as the head-tag triplet engines assemble into the consumer-side preview, plus the [direct answer](/direct-answer) lede (same job in the body). The layered place is in [the five layers of AEO](/five-layer-aeo); the catalog of failure modes is in [AEO problems](/aeo-problems).