Let’s be honest, International SEO and its tactics haven’t changed a great deal since I started doing SEO back in 2015. As an industry, we’ve gotten better at it and some very public misfires have mostly warned us off laziness and firmly into team “transcreation over translation”. Even the market share of search engines globally hasn’t really shifted, signaling that, on the whole, they’re doing a good job of serving the most relevant content to users.
Table of Contents
Key developments in international & localized search
- December 2011: Hreflang attribute – Introduced by Google, the hreflang tag allows webmasters to indicate the language and geographical targeting of a webpage, helping search engines to serve the correct language or regional URL in search results.
- August 20, 2013: Google’s Pidgeon Update – This update aimed to provide more accurate, relevant, and useful local search results by improving their distance and location ranking parameters. It increased the importance of local SEO by integrating more traditional web search signals into the local search results.
- June 2014: Google My Business – Replacing the ill-conceived Google+ and Google Places, GMB gave businesses easier access and control over their Local SEO and area-specific visibility.
- September 1, 2016: Google’s Possum Update – This update aimed to diversify local search results and limit spam, causing the rankings to fluctuate for businesses with an address outside the physical city area. This was a huge update (and headache) for many Local SEOs.
- December 9, 2019: BERT Worldwide – Google announced the worldwide rollout of BERT, including numerous languages. This update was significant for international SEO as it aimed at improving the understanding and interpretation of search queries across different languages.
- January 27, 2023: Yandex leak – Yandex, the leading search engine in Russia and the fourth-largest globally, experienced a substantial data leak, revealing approximately 17,800 ranking factors.
What International SEO looks like today
International SEO poses unique challenges due to language barriers, cultural differences, and search engine algorithms specific to each country. Currently, businesses rely on manual translation and localization efforts to optimize their content for different regions. However, this process can be time-consuming and may not always yield the desired results. Generative AI offers a solution to these challenges by automating and improving the international SEO process.
As the world’s leading search engine, Google has made significant advancements in understanding and ranking international or multilingual content. The search giant recognizes the importance of providing relevant and localized results to users worldwide. Here’s an overview of how Google currently approaches international SEO and ensures accurate rankings for multilingual content:
Language Detection and Localization
Google employs sophisticated algorithms to detect the language of web pages and understand the target audience for different regions. It also utilizes various signals, such as HTML tags, language declarations, and content analysis, to determine the language of a particular page. This language detection helps Google deliver search results that match the language preferences of users in different countries.
Country-Specific Search Engines and Indexing
Google maintains separate country-specific search engines, such as Google.co.uk for the United Kingdom and Google.fr for France. These search engines serve localized search results and prioritize content that is relevant to specific regions. When users perform searches on these country-specific search engines, Google’s algorithms consider location-specific factors to present the most appropriate results.
Google also indexes and crawls multilingual content across different regions. It takes into account the language and location signals to ensure that the right content is displayed to users based on their search queries and location.
Hreflang Attribute
To assist website owners in optimizing their international or multilingual content, Google introduced the hreflang attribute. This allows webmasters to specify the language and country targeting of their web pages. By implementing hreflang tags correctly, website owners can help Google understand the language and country relevance of their content, ensuring that it appears in the appropriate search results. International SEO guru, Aleyda Solis, has some excellent resources on hreflang tags.
Content Localization and Translation
For successful international SEO, it is essential to localize and translate content effectively. Google encourages website owners to provide high-quality translations that accurately convey the original meaning of the content. Using automated translation tools or low-quality translations may lead to misunderstandings or mistranslations, which can negatively impact search rankings.
Google’s algorithms analyze the quality of translated content and consider user feedback to determine the accuracy and relevancy of multilingual content. Website owners should focus on producing localized content that resonates with the target audience, adheres to cultural nuances, and aligns with local search intent.
User Signals and Engagement Metrics
In addition to language and location signals, Google considers user signals and engagement metrics when ranking international or multilingual content. Factors such as click-through rates (CTRs), bounce rates, time on page, and user feedback contribute to understanding user satisfaction and relevance of the content.
Understanding Generative AI
Generative AI might seem like the new kid on the block, but it’s actually been brewing for a while. Its roots trace back to the 1950s, but the 80s and 90s saw the rise of sophisticated models, like recurrent neural networks (RNNs) and generative adversarial networks (GANs). However, these models were held back by the lack of computational power and ‘big data’.
Things started to step up a gear in the 2010s, with the tech industry’s access to cloud computing and large datasets. This led to the emergence of advanced models, such as transformers, and more specifically, models like GPT-2 and GPT-3 developed by OpenAI. These models revolutionized generative AI, enabling the creation of human-like text and conversations. Now, as we’re deep into the 2020s, we’re witnessing the dawn of an era where generative AI isn’t just a novelty, but a game-changer in many industries, including international SEO.
With generative AI, SEOs are now able to scale up content production and workflows faster and more effectively than ever before. Creating content briefs, outlines and even first drafts with it can be a huge time saver and, depending on the quality of your prompt and editing, can produce excellent results.
Overcoming Challenges and Ethical Considerations
Generative AI certainly carries a lot of promise for international SEO, but it’s not without its challenges and ethical questions. One of the top priorities for businesses leveraging this technology will be ensuring the AI-generated content is both accurate and reliable, and importantly, it has to respect local sensitivities and cultural nuances.
We can’t turn a blind eye to things like AI bias. Depending on the data that the model is trained on and the backgrounds of those developing it, AI chatbots can produce offensive and sometimes even dangerous responses.
It’s important to also consider the prevalence of English as the web’s dominant language. In an article on Wired.com, Computer Scientist Thien Huu Nguyen said, “One of my biggest concerns is we’re going to exacerbate the bias for English and English speakers…” A real concern is that AI platforms such as ChatGPT are far less capable in non-English languages – it has been noted to be more likely to fabricate information and results when conversing in other languages.
By building AI capabilities and models predominantly on English-language sources and dialogue, we could certainly be creating a bias towards Western ways of thinking, and negative stereotypes held by such cultures can and have worked their way into AI responses.
It’s crucial to keep things balanced and fair, ensuring that no inadvertent preferences are sneaking into our content strategy. Transparency, too, is critical. People should be able to understand how the AI is working and making decisions.
And of course, there’s the accountability factor. When we’re relying on AI to steer the ship, we must remember that at the end of the day, humans need to take responsibility if things go awry. So, while the horizon looks bright with generative AI in the mix, we’ve got to tread carefully and ensure we’re using this tech in an ethical, responsible way.
Predictions and Observations on the Future of International SEO
The global market is massive, and AI is set to skyrocket our ability to reach it effectively. Let’s talk about how generative AI will give international SEO – and worldwide content accessibility – a much needed refresh.
- Go Local, Globally
Advanced AI won’t just translate content; it’ll make it locally relevant, adding regional flavor and cultural context. Your brand will be able to speak like a local from down-town LA, understand the context behind the great roll/bap/cob-type debates, and even interpret memes that would otherwise have needed research to decipher. I still feel like human intervention will be needed to expertly craft the relevant prompts, quality check and refine AI results, but I predict that a lot of the painstaking and manual work that international SEOs have to deal with today will shrink dramatically.
- Beyond Keywords
AI will lead us from simple keyword searches to complex, intent-driven searches. The result? More precise, context-sensitive search results that boost content visibility no matter the language or region.
Voice is the Future: With more people using voice commands, AI’s increasing knack for understanding diverse languages, accents, and speech patterns is a game-changer. Content accessibility just got an upgrade, even for those who love to talk to their devices.
- Visual Content Gets a Boost
AI’s ability to interpret images is getting seriously impressive. We’re talking about technology that can identify not just basic objects, but also complex scenes and even emotions expressed in images. AI can now describe the content of an image with surprising detail and accuracy.
This development has enormous implications for the world of SEO, particularly when we talk about “alt text” for images and translating them for global audiences (and targeted Image Search). Alt text is a description that’s coded into the image on a webpage, helping search engines understand the image’s content. In turn, this boosts your site’s accessibility and SEO.
Today, creating alt text is typically a manual task, requiring hours of human effort to write out an appropriate description for each image. To then create optimized alt text in each language you target significantly increases the effort required. With advancements in AI, we will be able to automate alt text generation. This means AI tools scan your images, understand their content, and automatically generate accurate, relevant alt text in any language. That’s a huge time saver and a fantastic way to ensure your images are SEO-friendly and accessible.
In fact, some tools are already popping up with this type of capability – see https://alttext.ai/
- Content Checks
Imagine if your AI could give feedback on content quality? That’s where we’re headed. Future AIs will assess relevancy, readability, and keyword usage, ensuring your content shines across different markets.
AI has really come a long way, especially in translation and content assessment. But let’s be real—it’s not perfect. Sure, it’s quick and reliable for dishing out translations in a ton of different languages. But even though it gets the job done, it can sometimes miss out on the subtle stuff like cultural hints or slang that’s specific to each language.
Think of AI translation tools like super-smart pattern detectors. They pick up patterns in huge amounts of text to guess what the translation should be. It’s pretty cool, right? But even then, they can get stumped by the trickier parts of language—like deeper meanings, idioms, and phrases that are unique to certain situations. That’s the kind of stuff a human translator can excel at.
Even with AI’s impressive translation skills, we still need a human touch right now—especially when it comes to making sure everything is ready to ship. And that’s where native speakers come in. They’ve got the cultural knowledge and language finesse that AI is still trying to get the hang of.
This human factor is super important for international SEO. When you’re trying to connect with local audiences, you need to really get their search intent, the way they use keywords, and their cultural context. Without these insights, your SEO campaign might fall flat. So even though we’ve got this amazing tech, remember that it’s still the combination of AI and human collaboration that makes the real difference.
- User Habits Uncovered
As AI gets better at understanding and predicting user behavior, we can tailor our SEO strategies to different cultural contexts and user preferences with more precision. We’re not just talking about translating content anymore. We’re talking about getting the right content, at the right time, to the right people. And doing it all in a way that respects their cultural contexts and personal tastes.
AI can help us work out the best time to post content. Even now, with ChatGPT-4’s code interpreter feature, it can analyze user data to figure out when people are most likely to be online and engaging with content.
Then there’s the type of content. AI can help us understand what’s really going to grab our audience’s attention. Is a flashy infographic the way to go, or would a long-form blog post be better? It can even give us insights into what sort of call-to-action (CTA) will get the best response based on the data provided.
And speaking of length, AI can help with that too. It can look at user data to tell us if our audience prefers bite-sized info or more in-depth articles. It can also tell us the sort of language and tone our audience prefers. Formal? Casual? Somewhere in between? The AI has got it covered.
So as AI gets smarter, we’re going to be able to fine-tune SEO strategies like never before. It’s going to give us a way to really connect with audiences around the world in a way that’s just right for them.
So, will I even have to translate my content in the future?
By developing language models like Lambda since 2021 and training them on dialogue instead of simply reading text, Google are gearing up to ‘translate the web’. In recent demos and talks, they have heavily hinted at someday being able to accurately provide content to any searcher, regardless of its original format or language.
Now, there will always be a need for signals that will ensure users aren’t served with irrelevant content. It’s all well and good translating a lawn service website into Spanish if that’s your preferred language, but if they don’t operate in your area then the content is practically useless. The signals we use to tell search engines about the areas we’re trying to target can include;
- Hreflang tags
- Server location
- ccTLDs
- Languages used
- Sub-folders
- NAP data
- Structured data
But, in the future, this list may be much less technical and more ‘obvious’ and ‘human’. For example, Google may be able to understand by your company phone numbers that you only operate in certain areas. Or, use data from reviews to pin point a rough service area. This is all likely for instances where a physical location is essential understand to fulfil intent, however it’s likely to look much different for informational queries and content that can be of use to anyone in the world.
Imagine you have a blog post on how to set up a CRM system, but the article is written only in US-english, uses dollars for currency and discusses a couple of example CRMs that only operate in the US. Perhaps Google will be able to still serve this content to someone in France by using AI to translate the content into French, changing the currency mentions to Euros and switching out example US only CRMs for global/French-focused platforms.
It’s exciting and a little nerve-wracking to think about how far search engines and AI companies could take internationalization. I for one am keeping a close eye on new developments and hope that, at the very least, the web becomes a far richer and more accessible place for all.
AI for International SEO in the wild
The fusion of AI with international SEO isn’t just a theoretical concept; it’s actively being employed by various companies and platforms to enhance global outreach. Here’s a look at how AI is driving the international SEO strategy in real-life scenarios:
Automated Language Translation Services
Companies like DeepL and Google Translate utilize advanced neural networks to translate text between languages. These AI-powered engines can detect subtle nuances and idioms, offering translations that are closer to human-like understanding. Though not perfect, they’ve become an essential tool for businesses to quickly translate content for international markets.
Cultural Adaptation with AI-driven Localization
Platforms like Unbabel blend AI with human expertise to offer localized content that’s not just translated but adapted to the cultural norms and expressions of the target audience. This AI-driven localization ensures that the message resonates with the local audience, enhancing engagement and conversion rates.
SEO Performance Analysis Across Different Regions
Tools like Pro Rank Tracker have incorporated AI to provide insights into international SEO performance. By analyzing rankings and visibility world-wide, they enable businesses to tailor their SEO strategies to the local market dynamics and keywords.
Content Generation and Optimization
Companies like MarketMuse utilize AI to analyze the content landscape across different regions and languages, suggesting optimization strategies and even generating localized content. This not only accelerates the content creation process but ensures that the content aligns with the local search intent and trends.
AI-driven Image Recognition for Global Audiences
As discussed, platforms such as AltText.ai and Cloudinary are leveraging AI to automatically generate alt text for images in various languages. This enables better image search optimization, making visual content more accessible and SEO-friendly across different regions.
Intelligent Chatbots for Global Customer Support
Many businesses are deploying multilingual AI chatbots like Ada and conversational AI software like Uniphore to provide seamless customer support across different countries. These chatbots can converse in various languages, providing personalized and instant support, thereby enhancing user experience and brand loyalty.
Final Thoughts
Generative AI presents immense opportunities for businesses looking to excel in international SEO. By harnessing the power of AI, companies can create multilingual content, optimize keywords, enhance user experience, and gain valuable insights to drive their SEO success. In the future, it remains to be seen whether content will still need to be localized by publishers, at least to the extent we’re used to. However, I feel that there will always be a place for native speakers and inhabitants in the way brands market globally – after all, they put the Expertise in E-E-A-T! If anything, we should all be advocating for more on-the-ground involvement from communities, developers, engineers and marketers from every corner of the world in AI- ensuring search truly serves and speaks in everyone’s language.