logo-small

13 Most Common Hreflang Mistakes - SEMrush Study

66
Wow-Score
The Wow-Score shows how engaging a blog post is. It is calculated based on the correlation between users’ active reading time, their scrolling speed and the article’s length.
Learn more

13 Most Common Hreflang Mistakes - SEMrush Study

This post is in English
Elena Terenteva
This post is in English
13 Most Common Hreflang Mistakes - SEMrush Study

Technical SEO mistakes can result in not only a bad user experience for your website, but eventually, your rankings, online visibility, and traffic will suffer. When it comes to international SEO, it turns out that hreflang implementation mistakes, in particular, can be extremely harmful. We want to prove this to you with our new research on the most common hreflang implementation mistakes.

The results of this research were quite expected, yet at the same time, they surprised us a lot. Let me explain:

If you’ve ever dealt with hreflangs, you know how many details need to be considered and how many pages need to be analyzed. During our research, we found out that the average multilingual website has around seven language versions; so we are talking about websites with a vast amount of pages here. Naturally, we expected that the frequency of mistakes would be high.

What really surprised us was that the number of hreflang mistakes would be so high!

Last year, the SEMrush team conducted research on the most common on-site SEO issues, which led to a huge debate: “Does it matter how many times I have one or another small mistake on my website--for example, a missing ALT attribute--if it has nothing to do with my rankings whatsoever?”

Okay, missing ALT tags can be forgiven, but hreflang mistakes, no matter how ‘small’ they are, can’t! Your goal is to reduce your bounce rate and improve your conversions by ensuring that your target audience lands on the version of your page that is most suitable for them.

However, without proper hreflang implementation, you are unable to clearly indicate to Google which page to index and display for a particular region. These factors have a direct impact on your visibility and traffic in the region.

It’s quite surprising to see that the percentage of websites with the simplest hreflang implementation issue - “wrong country code in a hreflang value”--is the same as the percentage of websites that have missing ALT tags on at least one page! But the consequences of these mistakes are dramatically different.

Therefore, we hope this research will call attention to hreflang implementation issues and help you to understand and estimate the potential danger of them. Also, we want to give you a hint as to which mistakes on your website should be fixed first.

How did we collect all this data?

We analyzed 20,000 websites that have multiple language versions using the SEMrush Site Audit tool - a powerful site auditor that checks websites according to approximately 50 parameters.

The SEMrush Site Audit tool offers 13 checks for hreflang implementation and is developed based on Google’s standards. Below you’ll see the percentage of websites with hreflang mistakes and issues which were detected on at least one page.

SEMrush research about hreflangs

1. Issues with hreflang values

Hreflang values seem to be the easiest part. All you have to do is use ISO 639-1 format for the language code and ISO 3166-1 Alpha 2 format for the country code. But these mistakes can be easily made by accident, even if you are very well versed in the subject of hreflang implementation. No one is immune to making a typo. You can use the hreflang tags generator tool by Aleyda Solis to prevent any mistakes.

There is a variety of these codes, so we decided to split them into 5 different categories.

“Issues with hreflang values appear on 15% of multilingual websites”

Unknown language code

The language code in your hreflang attribute must be in ISO 639-1 format. Otherwise, Google will ignore your hreflang attribute. Abbreviated terms we’re used to in everyday life are not working here. For example, “eng” in a hreflang value would be a mistake, “en” should be used instead.

Unknown country code

There is another format for the country code in a hreflang attribute--ISO 3166-1 Alpha 2.

The same - familiar abbreviations are not working in hreflangs. “UK” for United Kingdom? Nope, the correct hreflang code is “GB” (for Great Britain). So, if you are not sure which code to use, it’s best to double check.

Underscores used

Using underscores to separate language and country values in a hreflang attribute is not valid. Developers and webmasters use underscores a lot, but hyphens should be used instead.


Only country code is used

It’s Google’s standard that when using a country code, you must always provide a correct language code as well. Country code alone is not valid.

Invalid order of hreflang values

We also found that in certain cases, values in hreflang attribute are specified in reverse order. A language code must always precede a country code. It’s not always the typo, but it could come from a misunderstanding of Google’s basic requirements. Fortunately, it’s rarely made.

2. Hreflang conflicts within a page source code

In this section we collected hreflang implementation issues related to the conflicts in a page’s source code. They are difficult to detect manually because to identify them the entire page code should be examined, not just a single line of code or hreflang value.We detected these issues on 58% of all analyzed websites.

“58% multilingual websites have hreflang conflicts within a page source code”

No self-referencing hreflang tag

This is the most common issue in this section. We found out that if a website has a conflict within a page source code, in 96% of cases, the page doesn’t contain a self-referencing hreflang in its set of hreflang attributes. That means that those attributes may be ignored or interpreted incorrectly.

To fix this mistake, make sure to include the page’s URL and language code in your set of hreflang attributes.

Conflicting hreflang and rel=canonical tags

When using canonical tags on a webpage along with hreflang attributes, you should make sure to specify a self-referential canonical tag.

More than one URL is specified for the same hreflang value

If more than one URL is specified for the same language and country, Googlebot will ignore all of them because there is no clear message about which page should be indexed for this language version.

To fix this mistake, remove all conflicting hreflang URLs and make sure that you only have one URL specified for a particular language and country in your hreflang attribute.

3. Issues with incorrect hreflang links

The second group of mistakes is related to incorrect hreflang links. You have to point Google in the right direction to index all your pages. If the hreflang points to the page that doesn’t exist (4xx HTTP status code) or redirected, or you are using a relative URL instead of an absolute URL, these pages could be indexed incorrectly, or not indexed at all.

“37% of multilingual websites have issues with incorrect hreflang links”

Hreflang link points to redirected page

If your hreflang link points to a page that is redirected, the new page may not get indexed or even appear in search results.

Broken hreflang link

If your hreflang link points to a URL that returns a 4xx or 5xx HTTP status code, obviously it will be ignored by search engines.

Relative links

Relative URLs can be misinterpreted by Googlebot, so they might not be indexed.

Try to avoid them and give crawlers the correct path.

4. Potentially missing hreflangs

When you have a multilingual site, it’s very easy to forget to specify language for some of the pages.

We noticed that about 32% of the websites have different sets of languages for different pages. We understand that some hreflangs may not be specified on your pages deliberately, but this is not true in all 32% of cases.

If you want to double check that all your pages have required hreflangs, you need to check all of the pages manually. However, the SEMrush International SEO report enables you to check your website in a glance.

5. Hreflang language mismatch

When you specify a page’s language, it’s very easy to make a typo or use a wrong language code, thereby giving Google a wrong signal. For example, if you write a “fi” instead of “fr”, Googlebot will think that this page (written in French) must be shown to users who speak Finnish.

During the research, we saw that on 21% of all the analyzed websites page’s language, value is different from the detected page’s language (at least on one page).

Conclusion

As you can see, the majority of hreflang mistakes have pretty serious consequences: they can lead search engines to a dead end, therefore, confusing them by preventing pages and even URL sets from being indexed. We hope that this data will help you to begin improving your international SEO and avoid making new mistakes in the future. Keep your website healthy and don't forget to run a site audit!

We’d love to hear your comments about what kind of hreflang implementation mistakes you are dealing with most of all! Which ones cause the most difficult problems? Please share your experience!

Go global!

International SEO Audit with SEMrush

Please specify a valid domain, e.g., www.example.com

Elena Terenteva, Product Marketing Manager at SEMrush. Elena has eight years public relations and journalism experience, working as a broadcasting journalist, PR/Content manager for IT and finance companies.
Bookworm, poker player, good swimmer.
Share this post
or

Comments

2000 symbols remain
Marcin Marc
Do you have any knowledge how works itself tag:
EN
and like tags:
en-GB
en-AU
en-CA
en-NZ
What better way to influence website traffic?
Hi Elena, very nice article to better explain HREFLANG Issue. I'm still a little bit confused about Self-Referencing Hreflang Tag with canonical url. From your article it seems we shouldn't use both but I've never read a Google Article about this specific issue. Could you please give me a Google link about this?
Elena Terenteva
Francesco
Hi Francesco! Actually we didn't state it. Hope this article will help you - https://www.semrush.com/blog/7-common-hreflang-mistakes-and-how-to-fix-them/
"The hreflang tag also can be used along with rel="canonical" annotations, but hreflang tags need to reference self-referential canonical URLs. For example, page A should have a canonical tag pointing to page A, page B should have a canonical tag pointing to page B, and page C should have a canonical tag pointing to page C. All three pages should have hreflang tags that mention all three of the pages in the group. You do NOT want to canonicalize only one version of a page in a page grouping, as that would interfere with hreflang annotations."
Great article Elena! Have a question about the case "Conflicting hreflang and rel=canonical tags". How do you think search engines process this type of mistake? Are they ignoring all hreflang tags on that page or they are ignoring all hreflang tags on the website?
Elena Terenteva
Angel Stoyanov
Thank you so much!
Without self-referrencing hreflang you are giving Google controversial signal, asking to index and not to index all set of hreflang URLs at the same time. Conflicting hreflang and rel=canonical tag will affect the page and it’s language versions, but not the entire website.
Alex Lyubenkov
Hi Elena, great post! I'm curious why didn't you mention x-default value?
Elena Terenteva
Alex Lyubenkov
Hi Alex! X-default value is optional and absence of this value on the page can’t be considered as the mistake, so that’s why we don’t have any checks related to it. And actually, during the research, we found out that it’s rarely used - just 1% of multilingual websites have it.
Cheers
Alex Lyubenkov
Elena Terenteva
I see your point, thank for the research and quick answer!