Web Analytics

Translation Depth vs Breadth – which languages and which content?

Authors

Tim Branton

Tim Branton

PureFluent CEO

Ian Gilchrist

Ian Gilchrist

PureFluent Roving Reporter

Share this

Tweet Share Share

More content

  • Do translations drive ecommerce results even when customers speak some English?

    Read now
  • Is English ok for my customers?

    Read now
  • Should you add subtitles or voiceover to your video content?

    Read now
  • How is the gig economy debate affecting the translation business?

    Read now
  • How will Brexit affect the translation industry?

    Read now
  • Is the subscription model coming to the translation industry?

    Read now
  • When does it make sense for a company to do translations in-house?

    Read now
  • How long does a translation take?

    Read now
  • How much does translation cost?

    Read now
  • What to consider before you send content for translation

    Read now

8월 28, 2019

In this instalment of the PureFluent blog, Ian and Tim discuss translation depth and breadth: how much of my content should I translate, and into which languages?

Ian:This is such a “how long is a piece of string?” subject – what’s a good entry point for the discussion?

Tim:I think we could start with the concept of translation depth and breadth. You might for instance translate your core website into loads of languages but not translate your support material into all those languages, so you choose to focus on going as wide as possible but not worry about going too deep.

On the other hand you can take the opposite approach and decide that your core markets are the UK and Germany, so you’re going to take everything and translate it into German, the marketing content, support pages, your blog, etc etc.

I’m expressing this as an either/or choice between these options, and largely that is what you’re faced with. You don’t see people going both broad and deep, for the most part because it’s impractical. Someone might potentially be translating (and some do) into 50+ languages, and the time, cost and complexity of managing that level of translation is very difficult.

There’s an interesting study from CSA Research which looked at more than 1,000 multilingual websites, and you can see this graph really illustrates the point I’m making – the more target languages, the shallower the depth of translation.

©CSA Research 2019

So the reality is that people certainly tend to go one way or the other. The simpler the array of content that you have available for your customers, the more feasible it becomes in terms of both cost and complexity to go fairly deep.

Ian:It sounds like customers are faced with a tough decision, it can’t be easy to make that choice assuming you have a limited budget for translation?

Tim:For almost everyone there is some sort of compromise involved. Most of our customers are seeking a commercial outcome from their translations, so they’re looking for where the money is. Germany for example is a rich country with 80 odd million people, which you’re not going to ignore – if Germany is part of your plan you’re probably going to go fairly deep because of the substantial commercial opportunity.

If you’re considering the Netherlands, the decision is a tougher one – it’s another rich market, but only has a population of 10 million. Also, because it’s a small, wealthy country, its population has very good language skills, so there’s a weaker commercial argument supporting putting budget aside for translation, and Dutch is going to cost just as much to translate into as German. So, if you have to make this choice, you’re probably going to go shallower with Dutch than you would with German.

Potentially you have to take into account not only language, but also locale; if you’re translating into German, are you only going to translate into German for Germany or are you also going to translate into German for Switzerland and Austria? There will be solid arguments that encourage doing the translation variants and there will be arguments against doing so. If you do decide to translate into German for Germany, Switzerland and Austria you’re looking at not exactly tripling the cost, as there will be some efficiencies across the three, but it becomes complex, more expensive and time consuming.

Ian:Is this just a question of budget? I mean if I had an unlimited budget, would I go ahead and translate all my content into 20 or 30 languages?

Tim:It’s not just budget, you also need the ability to fulfil orders in all these markets. If you don’t have the right order fulfilment structure in place in a market, there is little point in translating for that market. Allied with this is another key factor, customer support. If you don’t have customer service available in a market’s language, you can translate your website and support materials into that language, but again you have to ask yourself is it worth doing so if you don’t have people who speak it within your organisation, meaning that you don’t have an active customer service capability in that language.

This isn’t a given though. For example, we’ve translated our web site into seven languages including Japanese and Korean, but our biggest markets are those of English speakers in the UK, US, Canada and Australia, and German speaking customers in Germany, Austria and Switzerland. While we’ve made our core web site available in seven languages, for the moment we’re making our blog available in English and German, which simply reflects where most of our customers are – we’ll get diminishing returns translating everything into everything.

Ian:So apart from customer support and order fulfilment, are there other things that would get in the way of adding more languages?

Tim:A key part of the translation process is ensuring that you have the right systems and infrastructure in place to handle multilingual content, otherwise you’re going to drown in the whole process. Most customers will have a content management system in place, they may have an ERP system and a learning management system (LMS) and a document management system like SharePoint, but the question remains, are all these different systems multilingual capable and can you plumb them into a translation management system. You can radically speed up the connection between your monolingual content and the translation process by integrating these systems, and it’s often not as hard as you might think.

It’s not just a question of budget: are your systems ready to handle multilingual content?

One more decision relating directly to your question “what languages shall I translate into?”: you don’t have to treat all your content in the same way. We spoke previously about machine translation and the different flavours of machine translation that one can look at, from very, very cheap on a marginal cost basis to post edited machine translation, all the way through to professional translation and trans-creation, whereby you’re essentially copywriting adapted content for a target market.

Again, it’s not just a question of “do you translate or not?” This decision making process is about how you translate and make intelligent choices about which type of content you translate, and how you manage that work flow creating multilingual content that you get out in front of your customers in a manner that delivers commercial benefits.

Ian:Now that you’ve explained the factors that should be applied to the “which languages do I translate into?” decision making process, the specific questions remain: Spanish? French? German? Italian?

Tim:I think the question comes back to thinking about market rather than language, where you see your product having the most potential. If you think your product is going to do particularly well in the Netherlands, it makes commercial sense to translate your content into Dutch.

Similarly, coming back to our previous conversation about Cantonese versus Mandarin, if you feel that you have a particularly strong offering that’s going to resonate with consumers in Hong Kong then your content has to be in Cantonese. I can’t stress strongly enough that your translation decisions must first and foremost be determined by potential in a market, not which country has the biggest GDP or the biggest population.

Ian:The Netherlands is an interesting choice of example as in my experience every single Dutch person speaks impeccable English, which would seemingly support the translation decision to not go very deep.

Tim:I can’t really argue with you on that! But that kind of thinking can result in a dangerous assumption that people regularly make, perhaps not about the Netherlands but about say Germany. A lot of people assume all Germans speak English, but it’s not true, and Germany is obviously a massive market, as are France and Spain.

Even if someone does speak English as a second language there’s a lot of evidence that shows that given a choice, and we all know the amount of choice consumers now have, customers will gravitate towards content in their own language. Unless you have a brand that’s strong enough to override a consumer’s natural preference to read content in her own language, and you do need to be very confident about that, and you must pay heed to the consideration of “where’s the money?” and what’s going to make it more likely that she will engage with your content and buy your product.

Given a choice, consumers will gravitate towards content in their own language.

I appreciate that this is a self-serving argument being put forward by a translation company, but there is really good research arguing for this and it isn’t just something which I’ve come up with.

Consider for example the performance of the most successful American companies, such as Microsoft and Apple who are the behemoths of the American economy, and the effort they put into translation. The sophistication of these companies’ approach to translation is leading the way, and not only in machine translation. If you examine the marketing literature for Pixel 4 in numerous languages, it’s beautifully written and I guarantee that machine translation went nowhere near those translations!

Google is of course a leader in machine translation but they are also making intelligent decisions regarding the translation of different content. The company knows that it’s not possible to use machine translation for everything, and they know that in order to be a truly global company they must have really good coverage for their content across a broad range of languages and that machine translation is only a part of the mix.

Coming back again to our starting question, there aren’t easy or straightforward answers: you can’t do everything, and there is also a danger of potentially falling between two stools. If you go broad and use up your budget covering 20 languages, you won’t have the funding you need for SEO and customer support in your core markets. Once again, there are always trade-offs between depth and breadth, but if you focus on the places where you think you have customers who are ready to buy your product, that gives you the best possible guidance as to where it’s worth making the investment in languages and how deep you go into them.

Ian:I imagine too that market evolution will play a role in translation decisions in some instances. As a company’s market penetration and market share grows there may be a convincing argument that the time has come for a company to translate more of their content.

Tim:Absolutely, that’s a really good point. You can put some key content out in a specific market in their language and see what level of response you get, as there are loads of analytics available that will tell you how your website is performing. If you put content out in 20 languages and find one’s doing better than you expected, that’s probably something you should heed and you should consider further translation investment in that market.

There’s a counter argument to this as well. You can get your website translated into let’s say Italian, but if you don’t keep your SEO in place, you don’t make your blog or social media posts available in Italian, and you don’t provide Italian customer service, when you assess your Italian performance and find it lacking, you then decide it’s not working there – but of course it’s not working!

Realistically, you have to factor all that’s lacking in a language translation into your assessment of how well you’re doing in a market. If you just do your website content in Italian and you’re getting a little bit of response, you then should ask yourself how much better you’ll be doing if you do your SEO and your blog and social media in Italian. It’s a complex area, obviously.

Our new website now includes content in Korean for the first time, and just the fact that we have that content in Korean is bumping us up in search results in Korea and creating traffic for us from there. Overall our traffic is completely global, and the main reason for this is that our content is in seven languages. If we were to add three more languages taking us to ten, we would significantly increase our audience.

It’s not just about the number of people looking at our website, as we built our website with commercial intent to grow our business, as are all our customers. The site has to provide a meaningful interaction for the people who visit it; we want them to take some form of action and of course we want to convert them into customers.

About the authors

Tim BrantonTim Branton

Tim Branton is PureFluent's CEO and a passionate advocate for the role of technology in the language industry. He has 30 years of business experience across the chemicals, telecoms, business services and software sectors in the UK, Singapore, Japan, China and South Africa.


See all posts by Tim Branton
Ian GilchristIan Gilchrist

Ian has worked in music and home entertainment product development, marketing, and journalism in the U.S., Canada and the UK, where he currently lives, for over 30 years.

In that time he's has aided and abetted an eclectic array of artists including Alison Krauss, Talking Heads, Madeleine Peyroux and Slade, and has worked for a diverse range of labels and companies including Universal Music (Canada), Pioneer LDC (Europe), Milan Records (France), the British Film Institute (BFI), Rounder Records Group (Canada) and BMG (UK). In his guise as a film journalist Ian's interviewed many renowned and influential people, including director John Carpenter (Halloween), actors Jesse Eisenberg (The Social Network) and Tom Hardy (Venom), director Roman Polanski (Chinatown), and many more.


See all posts by Ian Gilchrist

Share this

Tweet Share Share