Publishers that want to target international audiences often struggle with the right approach. Expanding your audience beyond your own country’s borders is challenging for a whole raft of reasons, and search is one of those.
I’ve consulted on dozens of international SEO projects for publishers. Most publishers have a preference for a specific approach, which some major news brands have adopted with decent success. However, this popular approach is not the ideal strategy for most publishers.
In this newsletter, I’ll share my experiences and show what you actually need to do if you want to succeed in search as a news publisher in multiple countries.
The popular approach: One generic TLD
Publishers look at sites like theguardian.com and believe that this is the optimal platform to build an international audience. The assumption is that with a generic .com domain (or any top-level domain that Google considers generic), a website can theoretically target almost any country. It works for The Guardian, so why not for you?
The truth is that a generic TLD for international audiences usually only works in specific contexts. The most important ingredient of a successful international strategy built on a single generic domain is brand power. The publisher needs to have strong brand equity, with large numbers of brand searches and direct traffic across all the territories that the site aims to target.
Additionally, a singular generic domain targeting multiple countries usually only works if the publisher uses a single language. Note that theguardian.com only publishes in English. While every territory targeted by The Guardian has its own subfolder homepage, individual articles on the site are not separated by country and have a singular canonical URL.
The moment you want to publish in multiple languages, the entire premise changes and a single generic TLD is rarely the best strategy.
Multiple languages on one domain
There is a context in which a news publisher can publish multiple languages on a single domain. This context is when your audience is in one country but could speak different languages. For example English and Spanish in the USA, or French and Dutch in Belgium, or any of the gazillion languages spoken in Switzerland.
In that case, you can publish translated versions of the same content on one domain. The moment you translate an article into a different language, it is no longer a duplicate and you don’t need to do anything special.
When an article is translated, it becomes its own unique piece of content. Google states it explicitly in its localisation documentation:
“Localized versions of a page are only considered duplicates if the main content of the page remains untranslated.”
When you translate a piece of content, for example an English article into Spanish, and publish it on a different URL, you don’t need to add hreflang tags or do any additional signposting. Google will recognise the language of the content and rank the article appropriately.
I still recommend making sure the HTML ‘lang’ attribute reflects the content’s language. Google doesn’t use it but other search engines do.
So, multiple languages on one domain can work, if it’s targeting users in the same country. What about when you want to target multiple countries and languages? First, let’s talk about multiple countries using the same language, like English-speaking users in the UK, US, and Australia.
Same language, multiple countries
The theory is that, when you have content in the same language targeting different countries, you can localise the content for these different audiences and use hreflang tags to point Google to your alternate URLs. Theoretically, this would ensure the right version will rank for the right users.
Unfortunately, this doesn’t always work. First, hreflang tags are hints, not directives, and Google doesn’t always obey them. Second, hreflang is messy and complicated to implement, and many implementations are invalid or incomplete.
And third, for news publishers there is one specific problem with how Google processes hreflang that makes it an entirely unviable mechanism. This has everything to do with how quickly Google indexes news content, and how slow it is to reindex already indexed articles.
It’s the same reason why we optimise articles for SEO before we publish.
Say you have an article in English targeting the UK, and you publish it on your website in the /uk/ country folder. The American desk then wants its own version of the article. They publish the piece in the /us/ folder on your site. Hreflang meta tags are added to both articles to tell Google that these are equivalent articles in different languages.
Let’s look at the timeline for this:
First, the UK-targeted British article is published. It is almost immediately crawled and indexed by Google. There is no translated version yet, so the article has no hreflang meta tags (or only EN-GB self-referential tags).
A short time later, US-targeted American article is published. It contains hreflang meta tags for both the EN-GB and EN-US versions. At the same time, the original article is updated with the same hreflang meta tags. Google will index the American article just as quickly.
However, now there is a conflict in Google’s index. Google sees the hreflang tags in the American article, which reference the British article. However, Google’s indexed version of the British article does not contain reciprocal hreflang tags for the American version.
You’d think that Google would use that as a signal to recrawl and reindex the British article, right? Wrong.
Google will probably prioritise recrawling of the British article to an extent, as the hreflang tags in the American article count as an additional reference, but this is not an instant process. It can take hours for Google to recrawl and reindex the British article, see its newly added hreflang tags, and then accept it as a fully reciprocal and valid hreflang implementation.
In the hours between, Google will not accept the hreflang tags on the American article as valid, because it hasn’t seen the updated hreflang tags on the British version yet. And in news, a gap of hours is an eternity.
Is there a solution to make hreflang tags work in the context of news? Yes, but it’s difficult and messy: You can change the original British article’s URL at the same time as you publish the American article (and ensure the hreflang tags in both reference the new URL). A new URL means Google will treat it as a new article, and index both.
However, this is not a guaranteed fix, is difficult to implement, causes potential problems with content duplication and internal redirects, and can be seen as an attempt to manipulate Google’s rankings. I don’t recommend it.
Alternatively you can accept cultural and language variations, and simply not republish articles in the same language for different countries. This is what The Guardian does; the site has various homepages for different territories, but articles always have just one singular URL.
Multiple languages, multiple countries
Now if you want to target multiple languages and countries, things can get more complicated. For only a handful of country and language combinations, you can probably get away with keeping everything on one domain with a subfolder structure.
In a scenario where you have two or more languages for one country, or one language for multiple countries, complexity arises quickly. Subfolders work to an extent, and you can also use subdomains to expand your international targeting. However, every such ‘solution’ that can work in its own small context introduces complexity and potential conflicts later down the line.
Eventually, you inevitably run into technical challenges, or the aforementioned hreflang issues, or the wrong country code subfolder or subdomain for a specific audience, and your international strategy will be compromised in one way or another.
The best approach: Country-code TLDs
There is one international SEO strategy that doesn’t suffer from any compromise, tends to have the best long-term results and the fewest technical complications, and aligns the most with individual countries’ cultural and editorial norms: Separate country-code domains for every country you want to target.
If you genuinely want to grow audiences in different countries and achieve long term success, you will need to embrace locality entirely. Don’t pretend you can target French election voters while sitting in an office in Berlin, or Italian La Liga fans while writing from a sports desk in Chicago.
International success is about localisation. This is more than translation: It’s about behaving like a local. Local language, local lingo, local cultural references, and local domains.
I’ve seen so many international SEO projects fail because they didn’t fully localise. Compromises were made for the sake of resources or efficiency. And those compromises always come back to haunt the project.
The most common compromises are technical. Investing in separate ccTLDs introduces extra overheads. But it is, without a shadow of a doubt, the one strategy that has the best chance of success.
People in France want to read stories on .fr websites. People in Italy prefer sports news from .it websites. While there are always exceptions to be found where some audiences don’t have strong preferences for a TLD, these are rare.
If you take your audience seriously, go to where they are. Launch a ccTLD in their country, with a local news desk, local reporting, and full local infrastructure for your news site. That way you are building the best possible foundation for your site’s success.
Keep in mind that when you want to grow an audience in a specific country, you are competing with websites that are fully based in that country. Your competitors are websites that have local domains, local brand power, local journalists, local sources and contacts, local everything.
If you don’t become just as local as your competitors, you’re always going to be at a disadvantage.
But, you say, this is hard. It’s complicated. It’s expensive.
Yes, it is. That’s exactly the point. If you want to succeed internationally, don’t half-ass it. Do it properly, or don’t do it at all.
NESS 2026
13, 14, 15 October 2026
We’re a month away from the 2026 News & Editorial SEO Summit, and it promises to be another awesome event.
Three days of talks and sessions, looking at everything from SEO in newsrooms, paywalls, Google Discover, AI and news, automation, growing your podcasts and video channels, winning on Reddit, and so much more.
It’s an online virtual event, which means you can view all the sessions and engage with the speakers and audience from anywhere in the world. Plus your ticket gives you access to all the session recordings to watch back whenever you want.
Go to newsseo.io to see all the speakers, the full agenda, and buy your ticket.
Miscellaneous
Official Google Docs:
Analyze your social and video platform content performance in Search Console
Help your readers find your site through preferred sources in Google Search
Latest in SEO:
How to Audit Publisher Websites in 2026 - Leadership in SEO
Did AI kill evergreen content? Let’s discuss. - WTF is SEO
How to Track Google AI Mode Traffic in Search Console - Suganthan
What Top Stories Inside AI Overviews Means For Publishers And Brands In 2026 And Beyond - SEJ
Technical debt is becoming an AI problem - Jono Alderson
These SEO Strategies Drive 90% of Your AI Visibility - Zyppy
Interesting Articles:
Google rolls out pay-per-value AI licensing program to publishers - Digiday
New research reveals AI answers only as trusted as the newsbrands they cite - Press Gazette
Google AI changes could deal further blow to publisher Discover traffic - Press Gazette
Surviving the clickpocalypse: What millions of search events say about AI eating the news - Scrunch
AI is eating your journalism. Here are the 11 tools you can use to take back control - Journalism UK
Publishers are rebuilding their newsrooms for the post-Google era - Digiday
The 2026 Generative AI Landscape Report - Similarweb
The Generative AI Learning Penalty: Evidence from Chinese Secondary Education - SSRN
Make sure to check out the latest episodes of Beers & SEO with Barry & Steve:
I was a guest on Search with Sean for a panel discussion with other news SEO luminaries Nicole Agius, Clara Soteras, Alison Moss, Jordan Mazza, and my podcast partner Steven Wilson-Beales. Watch the full discussion here:
Lastly, I also had an interesting 1-to1 conversation with Sean about the state of search in news, where we touched on a wide range of topics:
That’s it for another (and long overdue) edition. Make sure to share it with anyone you think would find it useful. As always, thanks for reading and subscribing, and I’ll see you at the next one.
The hreflang timeline is the part I'd never seen explained this clearly: by the time Google reindexes the original article with the reciprocal tags, the news cycle is already over. E-commerce has the same problem on a slower clock, so it just stays hidden longer.
"Localisation is more than translation" applies well beyond news, too. A lot of Brazilian brands trying to sell abroad translate the site and stop there, then wonder why nobody outside Brazil finds them.