SEO for International Websites: What Most Brands Get Wrong

SEO for international websites is the practice of structuring, configuring, and optimising a website so that the right version reaches the right audience in the right country or language. Done well, it compounds your organic reach across markets. Done badly, it fragments your authority, confuses search engines, and costs you rankings in every market simultaneously.

Most brands underestimate the complexity until they are already in the middle of it. The technical decisions made when you first expand internationally are hard to reverse, and the commercial consequences of getting them wrong show up slowly, which makes them easy to misdiagnose.

Why International SEO Fails Before It Starts

The most common failure I see is not technical. It is strategic. Brands decide to expand internationally, build out a website structure, translate some pages, and then wonder why organic traffic in the new market is flat six months later. The answer is usually that no one stopped to ask whether the target audience in that market actually uses search the way the home market does.

I spent a significant part of my agency career working across more than 30 industries and multiple geographies. One pattern repeated itself constantly: brands would treat international expansion as a replication exercise. Take what works at home, translate it, point it at a new domain or subfolder, and wait. It rarely worked. What works in the UK does not automatically transfer to Germany or Brazil or Japan, and not just because of language. Search behaviour, platform preference, competitive density, and commercial intent all vary by market in ways that a translation layer cannot fix.

If you are building a go-to-market approach that includes organic search as a core acquisition channel, the international SEO decisions you make are part of that broader strategy, not a technical afterthought. The Go-To-Market and Growth Strategy hub covers the wider framework for thinking about market entry and channel selection, and international SEO sits squarely inside that conversation.

The Architecture Decision That Shapes Everything

Before you write a single piece of content for a new market, you need to decide how your international pages will be structured at the URL level. There are three main options: country-code top-level domains (ccTLDs), subdomains, or subfolders. Each has real commercial implications.

A ccTLD approach means separate domains for each market: example.co.uk, example.de, example.fr. This gives the strongest geo-targeting signal and tends to build local trust with users. The cost is significant: you are building and maintaining separate domain authority for each property. Links to your .com do not automatically benefit your .de. You are essentially running multiple websites in parallel, each needing its own content investment and link acquisition.

Subdomains (de.example.com) sit in the middle. They are technically separate from the root domain in terms of how authority flows, though Google has historically treated them with some nuance. They are easier to host on separate infrastructure, which matters for large organisations with complex technical environments. But they share many of the authority fragmentation problems of ccTLDs without the geo-trust benefit of a local domain extension.

Subfolders (example.com/de/) are generally the right choice for most growing businesses. They consolidate all authority under one domain. Every link you earn, regardless of which market it comes from, contributes to the same root domain. The geo-targeting signal is weaker than a ccTLD, but you can compensate for that through hreflang configuration, Google Search Console settings, and content localisation. For businesses that are not yet dominant in their home market, splitting authority across multiple domains is almost always a mistake.

When I was running a performance-focused agency and we started picking up international briefs, the architecture conversation was always the one clients wanted to skip. They had already made the decision, or someone in IT had made it for them. Unpicking a ccTLD structure two years into an international programme is expensive and significant. Getting this right at the start is not a technical nicety, it is a commercial decision.

Hreflang: The Most Misunderstood Tag in SEO

Hreflang is the HTML attribute that tells search engines which version of a page to serve to users in a specific language or country. It looks straightforward in documentation. In practice, it is one of the most commonly broken elements in international SEO.

The implementation requires that every page in your international set references every other equivalent page, including itself. If you have an English UK page, a French France page, and a French Canada page covering the same topic, each of those three pages must contain hreflang tags pointing to all three, including a self-referencing tag. Miss one, and the whole cluster can malfunction.

The consequences of a broken hreflang implementation are not neutral. Search engines can serve the wrong language version to users, which increases bounce rates and signals poor relevance. In competitive markets, that signal compounds over time. I have audited sites where hreflang errors had been silently accumulating for 18 months, with no one connecting the flat international traffic to the misconfigured tags.

There are three ways to implement hreflang: in the HTML head of each page, in the HTTP header, or in an XML sitemap. For most CMS-managed sites, the sitemap approach is the most scalable and the least likely to introduce errors at the page level. Whatever method you choose, validate it. Tools like Screaming Frog will surface hreflang errors at crawl time. Build that check into your regular technical SEO process, not just your launch checklist.

Translation Is Not Localisation

This distinction sounds obvious when you say it out loud. In practice, most international content programmes ignore it entirely.

Translation converts words from one language to another. Localisation adapts content to reflect how people in a specific market actually think, search, and buy. The difference matters enormously for SEO because search intent varies by market, not just by language.

A practical example: the phrase “running shoes” in English UK and “running shoes” in English US might seem identical, but the search volume, competition, and purchase behaviour around those terms differs. Now extend that across a language shift, say from English to Spanish, and the divergence between what your home-market content says and what Spanish-speaking searchers are actually looking for can be enormous. A translated page that ranks well for the original keyword may not rank for anything in the target market because no one in that market searches for the translated equivalent.

Keyword research must be conducted independently in each market. You cannot reverse-engineer local intent from your home-market keyword list. This is where many brands underinvest. Proper localisation requires native speakers who understand the market, not just translators who understand the language. Those are different skills, and conflating them is expensive.

Understanding how search behaviour maps to commercial intent in different markets is also part of the broader question of market penetration strategy. Organic search is one channel within that, and its role varies depending on where a market sits in its development cycle.

Technical Signals That Support Geo-Targeting

Beyond hreflang and URL structure, there are several technical signals that reinforce your international targeting. None of them individually is decisive, but together they create a coherent picture for search engines.

Server location used to matter significantly. It matters less now, with CDNs widely used and Google’s crawling infrastructure more sophisticated. But if you are serving a German audience from a server in Australia with no CDN, you are creating an unnecessary performance gap. Page speed is a ranking factor, and it affects user experience regardless of how search engines weight it.

Google Search Console allows you to set a geographic target for subdomains and subfolders. This is a direct signal to Google about which market a section of your site is intended for. It does not override content signals, but it is a useful reinforcement, particularly in the early months of a new market build when you have limited local link equity.

Local link acquisition matters more than most brands acknowledge. A German-language page on your site will rank more effectively in Germany if it has links from German-language domains. This is not just about relevance signals. It is about the local web ecosystem recognising your presence as legitimate. Building that takes time and genuine market investment, which is another reason why treating international SEO as a quick translation exercise tends to produce poor results.

Structured data should also be localised where relevant. Prices, currencies, addresses, and contact details in schema markup should reflect the local market, not the home market defaults. It is a small detail, but it is the kind of detail that separates a properly built international presence from one that was assembled quickly and never quite finished.

When Organic Search Is Not the Right Channel

Earlier in my career, I was too focused on lower-funnel performance. I would have looked at a new market opportunity and immediately asked what the search volumes looked like. I have since learned to ask a different question first: is this a market where organic search is even a primary discovery channel?

In some markets, social platforms drive more commercial discovery than search. In others, marketplace search (Amazon, Alibaba, local equivalents) is where purchase intent lives. In others still, the competitive density in paid search is so high that organic rankings, even if achieved, capture a fraction of the available traffic because the page is below the fold of ads.

This is not an argument against international SEO. It is an argument for being honest about what it can and cannot do in a given market context. Organic search compounds over time, which makes it a valuable long-term investment. But it is a slow channel, and if you are entering a market where speed to revenue matters, building an SEO programme as your primary acquisition strategy is likely to disappoint.

The brands I have seen succeed with international SEO are the ones that treat it as part of a broader market-entry plan, not as a standalone tactic. They invest in it consistently over 12 to 24 months, they do the localisation work properly, and they accept that the returns are not linear. The ones that struggle are typically the ones that expected results in 90 days and pulled the investment when they did not see them.

Understanding how organic channels fit within a broader growth model is something BCG’s research on go-to-market strategy addresses in the context of financial services, but the principle holds across sectors: channel selection should follow audience behaviour, not internal capability defaults.

Duplicate Content and Canonical Confusion

One of the more persistent misconceptions in international SEO is that having similar content across multiple language versions will trigger a duplicate content penalty. It will not, provided your hreflang implementation is correct. Search engines understand that a French and an English version of the same page are not competing documents. They are the same content for different audiences.

Where duplicate content becomes a genuine problem is when you have multiple versions of the same page targeting the same language. An English UK page and an English US page covering identical content with no meaningful localisation differences is a problem. Search engines will struggle to determine which version to rank for which market, and you may end up with neither ranking well.

Canonical tags are sometimes used in international SEO as a blunt instrument to consolidate authority, but they should be used carefully. A canonical tag pointing from a localised page back to the home-market version effectively tells search engines to ignore the localised page. That is the opposite of what you want. Canonical tags and hreflang serve different purposes and should not be conflated.

The audit process for international sites needs to check for canonical conflicts specifically. It is one of those issues that is easy to introduce during a CMS migration or a template update and hard to spot without specifically looking for it. I have seen it appear on sites that had been technically clean for years, introduced by a developer who did not understand the SEO implications of a canonical setting change.

Measuring International SEO Performance Honestly

Measurement in international SEO requires more discipline than most teams apply. The temptation is to roll up all organic traffic into a single number and declare progress or failure based on that. That number is almost meaningless for understanding what is actually happening across markets.

You need to segment performance by market from day one. Organic sessions, rankings, click-through rates, and conversion rates should all be tracked independently for each target market. A strong performance in your home market can easily mask a flat or declining international programme if you are looking at aggregate numbers.

Ranking data is a useful leading indicator, but it is not a business outcome. I spent years on the agency side presenting ranking improvements to clients as evidence of progress. Sometimes it was. Sometimes we were ranking well for terms that generated no commercial value. The discipline of connecting ranking movements to traffic, and traffic to conversions, and conversions to revenue, is what separates a serious SEO programme from a reporting exercise.

Tools like Hotjar’s behavioural analytics can add a layer of understanding to international performance that pure ranking and traffic data cannot. Seeing how users from different markets interact with localised pages, where they drop off, what they engage with, provides the kind of insight that informs content investment decisions. It is not a replacement for search data, but it is a useful complement.

Brands that approach international growth with the same rigour they apply to their core market tend to find that organic search compounds effectively over time. The ones that treat it as a set-and-forget exercise tend to find themselves re-platforming or rebuilding international presences every few years, which is far more expensive than doing it properly the first time.

If you are thinking about international SEO as part of a broader market entry or growth programme, the frameworks in the Go-To-Market and Growth Strategy section are worth working through before you get deep into the technical implementation. The architecture decisions, the channel mix, and the content investment all need to be grounded in a clear view of what growth looks like in each target market.

About the Author

Keith Lacy is a marketing strategist and former agency CEO with 20+ years of experience across agency leadership, performance marketing, and commercial strategy. He writes The Marketing Juice to cut through the noise and share what works.

Frequently Asked Questions