What’s the Best Method of Translation? We Compare the Options.

If you’ve delved into the world of software localization before, you know that no two projects are the same. Each has it’s own requirements, audiences, and language nuances, so you need to be flexible when it comes to choosing the right localization approach.

Three localization methodologies stand out for most companies. As you choose the right one that fits your needs the best, you’ll need to ask yourself a few questions:

  • How big is my project? Is there a lot of overall content that needs translating and how fast do I need to get the project done?
  • Who are my new audiences for my app or website, and how fine-tuned do the translations need to be to accommodate them? Some apps target very sophisticated audiences that need professional-grade translation, and some don’t.
  • What are my resources? Large companies with pre-defined development and translation teams have a leg up on smaller companies with just a handful of developers and no localization budget. And if you have a huge user base (like in the hundreds of thousands or millions) you’ll hope to draw upon their collective support.

Once you map out some basic parameters, you can choose from one of three methodologies. Here’s how they each work:

Machine Translation

Automated localization platforms leverage a sophisticated machine translation approach to do most of the labor-intensive translation work up front. Users can simply plug content into the system and “pre-translate” content fast, allowing translation teams to focus on fine-tuning after the bulk of the work is done. Machine translation also leverages some great capabilities, such as usage rankings that translators have provided in previous projects to help you zero in on the right context, and in-app translation that allows you to see what translation strings look at in the context of the user interface.

Crowdsourcing

If you’re lucky enough to have a very large user base, why not make that volume a strategic advantage? Crowdsourcing opens up your translation project to your entire user base (Facebook and Twitter do it this way, but smaller sites can do it too). It’s not as easy to manage, so your project managers will need to be on their toes, but it does offer scale and variety. Crowdsourcing can also be good for more granular or targeted apps, where translators might not understand the professional nuance of particular string and will need to rely on the expertise of many.

Professional Translation

The good news is that no matter what you’re translating, there are plenty of professional language service providers (LSPs) and independent translators at the ready to support your efforts. LSPs are specialists in specific industries and languages so they can usually give you the best accuracy and the most fine-tuned results. And since LSPs usually lack the resources to handle the more technical aspects of internationalization, they work with localization platform providers and can be found en masse in their networks to generate business. With any luck, you can track down rankings for each translator to see how he or she has performed in past localization projects.

Is there one perfect approach? Not really, but you can use this overview as a guide to get your priorities, budget, resources and requirements all in sync.