With orders received from other non-English speaking countries and a desire to expand, many companies are faced with a choice of how far to go with the localisation of their site. 

New research helps brands make that decision, from a full localisation of site and all campaigns, to a partial localisation of just the keywords and ads.

As a multilingual search agency a question we’re often asked: should we translate our website or landing page if we want to sell overseas?

I’ve always thought the answer should be yes (for non-English speaking countries that is), since I wouldn’t consider buying from a German or French website unless I wanted something that I couldn’t buy from an English language website.

Why go to the bother of running the site through Google Translate to get a part-translation if there are alternatives in English? Nevertheless, it’s not uncommon to hear that ‘English is widely spoken in the Netherlands and Sweden, so we’d like to run campaigns there in English’.

There’s much talk of the UK economy needing to rebalance in favour of exports and of the opportunities a competitive exchange rate can offer, and with some companies unsure of their growth prospects from UK customers, the fact that companies are looking overseas makes a lot of sense.

So if you’re reading this considering whether you should translate your website, the European Commission recently published a Eurobarometer report revealing European online language preferences that might help you make that decision. (Flash Eurobarometer, Flash EB Series #313, May 2011).

The report found that “although 9 in 10 internet users in the EU said that, when given a choice of languages, they always visited a website in their own language, a slim majority (53%) would accept using an English version of a website if it was not available in their own language”.

So you might think that 53% isn’t bad, but note that 90% always visited a website in their own language when given a choice. If you’re competing in a market where websites are available in the local language, you’re instantly at a disadvantage.

Drilling further into the country data on this topic, just 22% of respondents from France strongly agreed with this statement compared to 66% of respondents in Sweden. This might be seen to reinforce that Swedish users are happy to use the web in English, but only when local language versions are not available.

More interestingly the report also presents data on the frequency of using a language, other than the respondents own, for searching or buying products or services. Here’s a summary of the data from a selection of countries:

Country All the time Frequently Occasionally Never
Germany 3.3% 10.4% 42.7% 43.0%
Spain 5.9% 17.7% 37.7% 38.0%
France 1.7% 12.5% 38.6% 47.1%
Italy 6.2% 19.6% 38.7% 33.5%
Netherlands 2.1% 13.0% 39.9% 44.8%
Sweden 3.5% 14.5% 42.7%


Note that Swedish and Dutch respondents are just as likely as the others to never search in English, and the percentage respondents that do so all the time is not any higher.

So if you’re still thinking about expanding into new markets, it looks like the level of English as a second language isn’t that significant. Perhaps you should factor in the translation of your site or landing page to really make the most of the opportunity?

I understand that the prospect of creating a fully localised version of an e-commerce website presents a significant investment, and one that should be properly considered, so there are several options that companies sometimes pursue to test the market:

  • Create a multilingual PPC campaign and direct it at your English site to assess the level of interest. If the site gets visits (but perhaps a low conversion rate) you can use the visit data to help you decide whether the investment is worth it.
  • Create a multilingual PPC campaign and a localised landing page (or microsite) before committing to a full-scale localisation.

We have run such ‘partial translation’ campaigns and note that they seldom, if ever, deliver good ROI. This is because they are normally up against local-language competitors, meaning conversion rates will be low. They should only be used to test traffic.

I realise there are many factors businesses need to consider when expanding overseas but maybe translation should be high up the list?