Things to consider before you internationalize (i18n) your website

You think about internationalizing your website? Great, but read this article to get the whole picture of it. Let's start by asking some questions about the need for a translated site. Most of the time we speak about a non-english site which should get translated to english.

  • Do i have products that are easily sellable outside of my country?
  • Do i have the resources to even sell outside of my country?
  • How much revenue plus will i get by offering my products to a wider audience?
  • Is the internationalization business or ego driven?

The answers to those questions should be very convincing in the first place to even start about thinking starting the i18n project.

The Problem of underestimation

Underestimating a problem can be good because otherwise some projects would have never seen the day of light but on the other hand, many projects fail because of it. This very topic is very easy to underestimate because automatically you just think about giving some non-english texts from your homepage to a translation office, paying some bucks and get back some text which some internet agency needs to link into the homepage.

Spoiler: I am writing this article mainly because i heavily underestimated the problem landscape. And i am guilty of not having answered the questions above for myself in the first place. If i would have done so, i would have done it anyways because it was more of a learning exercise anyway but for most companies, this is surely not the case.

But the problem is not the translation. That's most likely the easiest part of the whole project. I identified three major challenges on internationalization projects you must be aware of.

Performance Marketing / SEO

Performance Marketing

[Photo by Stephen Phillips - Hostreviews.co.uk on Unsplash]

A lot of companies don't care about Performance Marketing or Search Engine Optimization. For them it's easy to do a i18n project. They just submit their existing textual content to a translation agency, get back the translated text and somehow implement it into their homepage. Project done. But let's be honest, those are not the most successful companies anyway.

So let's focus on the challenges:

  • Your current content is highly keyword optimized, which was most likely done by a SEO agency or a general internet agency (or by us).
  • You current URLs are also heavily optimized towards performance marketing

Just translating the content will not cut it because this is not keyword optimized. A translation agency doesn't know what SEO is. So before translating the content, you need to do the same work you have done with your primary language. You need to:

  • Do a keyword analysis in the target language
  • Do a volume analysis in the target market (search region)
  • Do a complete new backlink strategy

After this you need to sync the translation process with the SEO process.

Content Creation

Content Creation

[Photo by Aleks Marinkovic on Unsplash]

This is hughe. And i tell you why. We all agree on this sentence:

Content is what drives people to your site if you are not paying for it.

So if you have a hard time producing quality content on a regular basis in your primary language, you won't see progress in that department by adding another language which means doubling the work to do. And Content is not only about text. If you really want to cover some ground here, you also want to have video on your site. This is even more demanding in a secondary language because of the production process being more complicated.

Bear in mind what i have written about Content in the context of SEO. This adds another whole dimension to the content game.

Technical Implications

Technical Implications

[Photo by Christina @ wocintechchat.com on Unsplash]

I will divide this chapter in a few subchapters because it's a wide field here.

Language detection

This a website development issue. You need a strategy to detect the language in the first place to kind of guess which language the user wants to see before even entering the site. There are some options available which can be implemented like:

  • top level domain decides the language
  • browser language decides the language
  • cookie decides the language (prior visit required - fallback if not)

Content Publishing

We have several challenges on the technical side too. First we need to have a CMS which can handle i18n. A lof of systems have a minimum featureset for this. But some of them already fail in providing the option of having different URLs depending on the language.

Analytics

You can't decide on a marketing strategy without measuring all kind of things right? Unfortunately it's also more effort to split the analytics data into different target groups. Perhaps you want to use a different analytics account with your current provider, perhaps you want to use one account and separate the data based on filter and aggregations. Whatever path you chose, its effort to do so.

Hosting / Distribution

The idea behind an i18n project is most likely to attract leads from different countries than your home country. This poses new challenges when it comes to hosting your site. It's no longer enough to place the server somewhere in your country to have acceptable response times, you need to place the content in different country / zones to have good response times for all. This is achieved by CDNs. That's by the way the reason why Netflix is able to stream worldwide without much glitches.

GDPR / DSGVO

Here we go again. Everybody loves this topic right? Just one note to this one. Your GDPR strategy needs to be overhauled because every country has a different approach to it. We in germany have quite a strict one when it comes to things like cookie banners and stuff. Of course you need to adapt all that depending on the audience. You really dont want to present a cookie banner for a US customer visiting your site, but this might depend if you have a US branch or not. I am not lawyer, but i am pretty sure this topic is certainly a time killer for a i18n project.

Summary

I will never advice against an i18n project. For bigger companies it's not a question anyways but especially for smaller companies, doing a boneheaded project without enough budget will even hurt your existing homepage with your primary language, resulting in revenue DECREASE overall.

I rolled back the okaycloud site a few days ago and just left the blog in english because of the big challenges in the SEO department. And to be honest, this consultancy is way too small to really go for international clients on purpose. There is so much market in my home country, the energy is better spent there. But the learnings were immense. I am now better prepared than ever to really consult on that topic, having struggled myself and see what needs to be done to do it right.

Every challenge mentioned is solvable, from a technical standpoint as well as from a production standpoint. But the effort one needs to put in is also remarkable. I hope i could outline a few things so that people in charge of starting projects like this, really think about the implications. At the end, the CEO of a well run company most likely want to see the benefits of the project and compare it with the costs. At that point in time you better pull off some nice analytics with some uptrending charts when it comes to visitors, leads and potentially customers.