Dark green Relative Marketing Group cover graphic headlined "Why Google Ignores Your hreflang Tags" with a diagram of en-US and en-GB return links, x-default and en-UK ignored.

A US visitor searches your brand and lands on the UK page, with prices in pounds. The site has hreflang tags on every page, so the assumption is that Google is ignoring them. Usually it is: not out of caprice, but because the tags break one of a small number of documented rules, and when they do, Google discards the annotation rather than guessing what you meant.

This post walks through how hreflang actually works, the specific ways implementations fail, and a check you can run on your own site today. Everything below is drawn from Google’s own documentation on telling Google about localized versions of a page and managing multi-regional and multilingual sites, plus the patterns we run into when auditing sites that serve more than one country or language.

What hreflang does, and what it doesn’t

hreflang is an annotation that says “this URL and these other URLs are the same content, localized for different languages or regions.” It does not translate anything, it does not redirect anyone, and it does not tell Google what language a page is written in. Google states plainly that it does not use hreflang or the HTML lang attribute to detect a page’s language. It determines language algorithmically from the visible content. hreflang only matters once Google already has several candidate URLs and needs to decide which one to show a particular searcher.

That distinction explains a lot of confusion. If your “Spanish” page is really an English page with a translated menu and footer, hreflang will not make Google treat it as Spanish. Google’s documentation notes that localized versions are only considered duplicates if the main content remains untranslated, and it recommends one language per page for content and navigation, without side-by-side translations.

The three ways to declare it, and why you should pick one

Google accepts hreflang in three places: <link rel="alternate" hreflang="..."> elements in the HTML <head>, an HTTP Link: response header (the only option for non-HTML files such as PDFs), or <xhtml:link> entries inside an XML sitemap. Google describes the three methods as equivalent and says using more than one has no benefit in Search.

In practice, running two methods at once is where drift starts. A plugin writes hreflang into the head, a separate process generates the sitemap, and six months later a new locale is added to one but not the other. Now Google sees two conflicting sets of alternates for the same URL. Choose the method your stack can keep accurate automatically, and remove the other. For most WordPress sites that is the head; for very large catalogs it is often the sitemap, because sitemap child entries do not count toward the sitemap URL limit and they do not add weight to every page.

Why Google ignores hreflang tags: the documented causes

When hreflang fails, it fails quietly. These are the causes Google lists, in the order we most often find them:

Missing return links: if page A lists page B as an alternate but page B does not list page A back, Google ignores the pair, a safeguard so another site cannot declare itself an alternate of yours.

Missing self-reference: each language version must list itself as well as every other version, so the set of tags is identical on every page in the cluster.

Relative or protocol-less URLs: alternate URLs must be fully qualified, meaning https://example.com/foo, not /foo or //example.com/foo.

Invalid region codes: the region must be an officially assigned ISO 3166-1 Alpha-2 code, and Google says reserved codes such as UK, EU and UN have no effect, so en-UK does nothing and the correct value is en-GB.

A country with no language: the first code is always an ISO 639-1 language, and Google does not derive language from a country, so hreflang="be" targets Belarusian speakers, not Belgium.

Unsupported composite codes: only ISO 639-1 languages and ISO 3166-1 Alpha-2 regions are supported, which means the common es-419 value for Latin American Spanish is explicitly not supported.

Tags outside a well-formed head: hreflang <link> elements must sit inside a valid <head>; a stray element that the browser treats as body content can close the head early and strand every tag that follows it.

The en-UK mistake, in one paragraph

The single most common hreflang error we find on English-language sites is hreflang="en-UK". It looks right, it passes a casual review, and Google ignores the region part entirely. Google’s international documentation names UK, alongside EU and UN, as a reserved code that has no effect in Search, because the ISO 3166-1 Alpha-2 code for the United Kingdom is GB. A site that labels its British pages en-UK has not told Google those pages are for the UK at all, which is why a UK searcher can still be shown the US version. Changing every en-UK value to en-GB is a two-letter fix that most sites can ship the same day.

Failure modes the documentation doesn’t spell out

The rules above are Google’s. The following are operational patterns we see in audits, where a correctly written template still produces broken hreflang:

Return links that point to redirects: a locale page gets a new slug, a 301 goes in, but the other locales still list the old URL. The annotation now points at a redirect rather than the final page, and the bidirectional pair is no longer clean. Treat hreflang URLs like internal links: they should resolve directly with a 200. Our guide to how Google reads 301 and 302 redirects covers why chains matter.

Canonicals that contradict the cluster: every regional page canonicalized to the US version tells Google the UK and Australian pages are duplicates of the US page, while hreflang says they are distinct alternates. Google’s guidance for same-language duplicates across regions is to pick a preferred version and use rel=”canonical” together with hreflang, so each localized page you want shown should normally carry a self-referencing canonical. Our post on why Google picks a different canonical than the one you set explains how that conflict is resolved.

Unpublished locales left in the set: a regional page is taken down or set to noindex, but its hreflang entry stays on every other page. The other pages are now advertising an alternate that cannot appear in Search.

x-default, and why auto-redirects undermine all of it

The reserved value x-default names the page for anyone whose language or region matches none of your versions. Google says it was designed for language-selector pages and works best there, and that it needs no language code of its own. If you have en-GB, en-US and en-AU versions, an x-default pointing at a country selector, or at your generic English page, catches a searcher in Ireland or Singapore.

Where many sites undo their own hreflang is the automatic redirect. Google advises against redirecting users to a language version based on a guess about their language, and against using IP analysis to adapt content. The reason is mechanical: Google says most of its crawls originate from the US, that it does not vary crawler location to look for variations, and that Googlebot sends requests without an Accept-Language header. A site that silently redirects “US visitors” to /us/ may be redirecting Googlebot away from every other version. Show a dismissible banner suggesting the local version instead, and let the hreflang do the routing.

A check you can run on your site today

Open any localized page, open the browser’s developer console (F12, then Console), and paste this line:

[...document.querySelectorAll('link[rel="alternate"][hreflang]')].map(l => l.hreflang + '  ' + l.href).join('\n')

It lists every hreflang alternate on the page. Then check four things. Does the list include the page you are on? Does every code start with a two-letter language, and is any region an assigned country code (GB, not UK)? Is every URL absolute and HTTPS? Now open two of the listed URLs and run the same line: the output should be identical on each. If any page’s list differs, that pair is the one Google is likely to be ignoring. For hreflang delivered in a sitemap or HTTP header instead, inspect the sitemap file or the response headers, because the console line only reads the HTML.

What hreflang can’t fix

hreflang is a routing hint between pages Google has already indexed and considers equivalent. It will not get an unindexed locale indexed, it will not override a canonical that points elsewhere, and it is not a ranking boost. Google also says geotargeting “isn’t an exact science,” so give visitors a visible way to switch region on every page for the times the routing still misses.

International setups touch templates, redirects, canonicals and sitemaps all at once, which is why we treat them as a technical SEO project rather than a plugin setting. If your site serves more than one country or language and searchers keep landing on the wrong version, our SEO services team can audit the full cluster and fix it at the template level.

×
✦ Meet MIA – Your AI Marketing Assistant

Get Instant Marketing
Answers + Book Strategy Call

Talk to MIA and get real answers, custom insights, and expert help to grow your business.
💬
Answer your marketing questions

Get instant expert answers to help you grow.

📈
Analyze your website or ads

MIA can audit and analyze performance in seconds.

💡
Get growth strategies

Receive tailored recommendations.

📅
Schedule a call with our team

Book a strategy consult instantly.

Start Voice Chat with MIA Click to talk
📅 Book a Strategy Call Schedule a consult
⚡ Get Quick Answers Fast responses
🔍 Request Free Audit Analyze my site
MIA Bot

MIA

● Online
Listening...
🎙️

Click the mic to start speaking. We only use your voice to understand your questions.

Skip to content