The Internet is known as the great equalizer — having a website means access to every single market in the whole world. In other words, as long as you have an online presence that can be accessed and understood by most, if not all, you can sell to anyone without being restricted by geography.
With 1.4 billion Internet users communicating in English, 1.4 billion in Chinese, and 5 billion in the other top 10 languages, English is the most common language used online. However, more than half of the 7.7 billion Internet users worldwide speak another native language.
This is why localizing web content to cater for these non-English native speakers is the most logical step for improving sales conversions. Reaching 7.7 billion people instead of a mere 1.4 billion means a whopping 450 percent growth. An online business needs a site that can converse and interact like a native speaker to make significantly more sales.
The thing is, localizing sites is cumbersome, time-consuming, and expensive. MT might be quick, but it doesn’t guarantee accuracy and cultural appropriateness.
Cited from Harvard Business Review, a study by Gallup on language preferences in 23 European Union countries revealed that 9 out of 10 Internet users always visited sites in their own native language. Also, 19 percent of Europeans have never browsed in a language other than their own, and 42 percent of them had never purchased products and services from online businesses written in other languages.
A recent CSA Research survey of 3,002 non-English-speaking consumers from 10 countries confirmed these findings. The survey found that most respondents preferred to purchase from sites written in their native language. Thus, localizing web presence is a necessity in this global Internet economy and no longer a luxury.
In this article, we’ll discuss why localization is a must, explore site translation options, consider the pros and cons of each choice, and help you choose the route to take for translation and localization.
First things first, what’s “localization” exactly? How is it different from “translation”?
In a nutshell, website localization is the process of adapting an existing website to the language and culture of a target market. Translation, on the other hand, is limited to translating an existing text into a different language, which might not include the adaptation of culture and other nuances.
Naturally, localization requires modifications to the source content to reflect specific linguistic and cultural preferences that are relevant within the target market. This might include changing of illustrations, images, color schemes, and the overall design while maintaining the website’s integrity and message. For instance, marketing new products to China would demand specific handling, as the Great Firewall requires all content to be both available in Mandarin and altered to cater for search algorithms and keywords in Chinese characters as well as Hanyu Pinyin.
Web pages that are translated without consideration of these requirements are destined to fail. And we haven’t even talked about how cultural localization should be reflected on web pages. Catering to the Chinese market differs greatly from serving western countries, as they can be conservative in certain areas but more extravagant in others. Chinese society is also more communal and less individualistic than the western civilization.
With many factors to consider, creating a second website or a foreign language section should be done with a lot of care. Adapting an existing site to a new target language and culture takes more than understanding what to do. More importantly, the project must use the right tool for the right job.
Second, what would constitute a “good” localization? And why is it crucial to a web presence?
Linguistically and culturally adapted websites would increase engagement with visitors by reducing their cognitive efforts in processing information and making the overall local user experience more favorable. Once they arrive at your site, they can immediately grasp the overall idea and understand what to do next without overthinking. Therefore, the more “natural” the site feels, the better the result.
Ideally, the localized site doesn’t feel translated at all. Rather, it should read and feel like it was written for the reader in their own native language. To optimize the result, content localizers should take into account the language, belief and value systems, customs, traditions, habits, and education level of all users
It goes without saying that well-localized sites are crucial to attracting organic traffic and increasing conversion. Imagine an e-commerce store that’s only visited by English-speaking customers. By being served in their own language and culture, consumers would feel a stronger connection, so they’d be more confident in their buying decisions. After all, global marketing requires a comprehensive understanding of strategizing locally.
Before you can properly localize any online presence, the first stage is translating the content, which should be performed carefully and step-by-step. However, the steps can be quite different depending on whether you choose the machine translation, human translation, or computer-aided translation (CAT) route.
Which one to select would depend on various factors, including purpose, timeframe, and the resources involved.
If you’re a reader and happen to encounter websites written in another language, most likely you’d use free online translators for convenience. Simply copy and paste the text and wait for a few seconds for the machine to come up with the translation in the target language. Or, you can use a web browser plugin. This is called machine translation, or simply MT.
If you own a website and are toying with the idea of creating multilingual sites, MT can still be used with post-editing, considering web page content is an important marketing tool. However, if you simply need to analyze content for basic comprehension and internal communication, without any post-editing, MT is an ideal choice.
Some of the most popular machine translation engines are Google Translate, Bing Translator, Yandex, and DeepL. One way for translating web pages is by embedding auto-translate plugins, such as the Microsoft Translator widget and the Google Translate plugin. Note that the latter is no longer supported by Google as most web browsers now come with a native translation feature. Thus, such a plugin might no longer exist in the future.
While using MT is super fast, inexpensive, or even free, machine translations can be inaccurate, as this type of translation is accomplished by feeding a text to a computer algorithm that is programmed to process automatically. No human is involved in the process at all.
With such limitations, MT is suitable for translation of materials with no cultural nuances. Machine translators also don’t recognize idioms and phrases. Thus, it wouldn’t work well for documents containing cultural and ethnic messages, legal agreements, or scientific and technical papers.
Professional translation is ideal for projects that must be linguistically accurate, culturally correct, and exquisitely nuanced. However, it takes a lot of effort and resources to launch a human-powered translation project.
Many independent freelancers and translation companies or language services providers (LSPs) are now competing for projects online, through either marketplaces or requests for proposals (RFPs). For customers who don’t have any prior experience with translation services, finding a pool of highly qualified translators can be time-consuming and intimidating.
The steps involved when using human translation include assessing readiness, determining the localization project scope, getting the logistics in order, and, eventually, launching the project. After the launch, it still takes a lot of effort to manage the project milestones and to post-edit the output for quality assurance. Luckily, there are tools that allow you to translate your website in an automated fashion. For example, Smartcat has connectors for Wordpress, Drupal, Bitrix, and Magento websites.
PEMT refers to human translators’ editing the machine translation output to make sure it is as accurate as possible. Sometimes — although not always and not necessarily — they’ll also give the translation a “human touch” by adding revisions that no machine is capable of.
In a light post-editing process, the translator makes small changes to increase text comprehension without improving the style and the flow. It’s generally used for communication with a short lifespan or for content confirmation. Full post-editing is needed to remedy poor translation outcome, to localize cultural nuances, or to improve the style and fluency.
A caveat to post-editing is that it requires the use of computer-aided translation (CAT) tools, which can be quite pricey for individual translators or LSPs to use. Luckily, some aren’t. For example, Smartcat is free for all users, including end customers, LSPs, translators, project managers, and other team members. Smartcat generates revenue from its marketplace, payment automation solution, and other value-added services, so the company can afford to provide the CAT functionality for free.
Whichever approach you choose, you’ll likely need an environment where you will manage all your translation projects, so that you don’t drown in endless email threads and file exchanges. As that’s a topic for another article, we’ll just quickly outline Smartcat’s features that will support you in this process.
First, Smartcat allows customers to select the most qualified and experienced translators from a large pool of freelancers or companies, without having to use other platforms. Second, with Smartcat, machine translation, human translation localization, and post-editing can all be executed from the same dashboard.
Finally, flexible project management features provide the convenience of working with remotely distributed teams and the ability to scale rapidly without having to worry about the number of team members. All project participants can work together simultaneously and be constantly updated of translation memories, glossaries, and quality assurance checks.
So, which approach would you choose for your own website and why? We’re looking forward to your comments!
in Smartcat blog , May 19, 2019