Meta Tags Cheat Sheet

Which head tags matter to Google, and which do nothing.

The head of an HTML page can hold dozens of meta tags, but Google Search only uses a short list of them. The title element and the meta description shape how a result looks; the robots and googlebot meta tags control whether a page is indexed and how it may be shown; and a handful of others handle verification, character encoding, translation and SafeSearch. Everything else is either used by browsers and social apps, or ignored by search entirely.

This cheat sheet follows Google Search Central's list of supported meta tags and attributes, plus the link elements that matter for indexing such as rel=canonical and hreflang. The ignored tags are worth knowing too: meta keywords has had no effect on Google rankings for well over a decade, and Google detects a page's language from its visible text rather than from lang attributes. Open Graph and X (Twitter) card tags don't affect rankings, but they decide how your links look when people share them.

Meta tags Google supports

Tag	What Google does	Good to know
<title>…</title>	The main source for the title link shown in results. Not a meta tag, but the most important tag in the head.	Google may rewrite it, often from the H1 or anchor text, when it's missing, vague or stuffed.
<meta name="description" content="…">	May be shown as the snippet when it describes the page better than text from the page itself.	Not a ranking factor. Write one per page; duplicates across pages are less likely to be used.
<meta name="robots" content="…">	Controls indexing and serving for every crawler that reads it: noindex, nofollow, nosnippet, max-snippet and others.	The crawler must be able to fetch the page to see it, so don't also block the page in robots.txt.
<meta name="googlebot" content="…">	The same rules as robots, for Google's crawlers only.	Where both apply, Google follows the more restrictive rule.
<meta name="google" content="notranslate">	Google won't offer a translation of the page in search results.	Use translate="no" on an element to keep just part of a page untranslated.
<meta name="google" content="nopagereadaloud">	Stops Google's text-to-speech services from reading the page aloud.	Rarely needed.
<meta name="google-site-verification" content="…">	Proves you own the site in Search Console.	Keep it in the head of the home page for as long as the verification is needed.
<meta charset="utf-8">	Declares the page's character encoding.	Put it in the first 1024 bytes of the document. UTF-8 is the right choice for new pages.
<meta http-equiv="refresh" content="0;url=…">	Followed as a redirect. An instant refresh is treated as permanent, a delayed one as temporary.	Prefer a server-side 301 or 308; not every browser or crawler handles meta refresh well.
<meta name="viewport" content="width=device-width, initial-scale=1">	Tells browsers how to scale the page on mobile devices.	Google indexes the mobile version of pages, so a page without it renders poorly for mobile-first indexing.
<meta name="rating" content="adult">	Labels the page as sexually explicit so SafeSearch can filter it.	Also accepted: RTA-5042-1996-1400-1577-RTA.

Link elements that affect indexing

Tag	What Google does	Good to know
<link rel="canonical" href="…">	Suggests the preferred URL among duplicates. A strong hint, not a directive.	Use absolute URLs, one per page, pointing at an indexable page that returns 200.
<link rel="alternate" hreflang="…" href="…">	Tells Google about language or regional versions of the page.	Every version lists every other version and itself; add x-default for a fallback.
<link rel="alternate" media="…" href="…">	Connects a desktop page to a separate mobile URL (m-dot sites).	The mobile page points back with rel=canonical.
<link rel="next"> / <link rel="prev">	Not used by Google as an indexing signal (confirmed in 2019).	Harmless to keep; browsers and other engines may use it.

Tags Google ignores

Harmless to keep, but they won't help you rank.

Tag	What Google does	Good to know
<meta name="keywords" content="…">	Ignored by Google Search, and has been since at least 2009.	Competitors can read it, so it mostly gives your keyword list away.
<meta name="revisit-after" content="7 days">	Ignored. Crawl frequency is decided by Google.	Use a sitemap with accurate lastmod dates instead.
<meta http-equiv="content-language"> / lang attribute	Not used to detect a page's language: Google uses the visible text.	Still set lang on <html> for accessibility and browser features.
<meta name="author">, generator, copyright	No effect on Google Search ranking or display.	Fine to keep for other tools.
<meta name="robots" content="noarchive">	No longer used by Google Search since cached links were removed in 2024.	Other search engines may still honor it.
<meta name="google" content="nositelinkssearchbox">	Obsolete: Google retired the sitelinks search box in November 2024.	Safe to remove.

Read by social networks and chat apps when someone shares your link, not by Google Search.

Tag	What Google does	Good to know
<meta property="og:title" content="…">	Title of the link preview on social networks and chat apps.	Not used by Google Search for ranking.
<meta property="og:description" content="…">	Description line under the title in many link previews.	Can differ from the meta description.
<meta property="og:image" content="https://…">	Preview image. Use an absolute URL; 1200 × 630 pixels is a widely supported size.	Many apps cache the image, so changes can take time to show.
<meta property="og:url" content="https://…">	The canonical URL that shares are counted against.	Match your rel=canonical.
<meta property="og:type" content="website">	The kind of object: website, article, product and others.	website is the safe default.
<meta name="twitter:card" content="summary_large_image">	Card layout on X (Twitter): summary or summary_large_image.	X falls back to og: tags for title, description and image.

Good to know

Where meta tags must go

Meta tags only count inside the <head>. If a tag lands in the <body>, for example because an invalid element closed the head early, Google may not read it. Check the rendered HTML, not just your template.

Titles and descriptions get rewritten

Google writes its own title link or snippet when it thinks yours doesn't describe the page well for the query. Descriptive, unique titles and descriptions are rewritten less often, but there's no tag that forces Google to use them.