Evaluating localization tools: formulating the right criteria

It can be very difficult to evaluate a localization tool and decide if it fits within your localization process and delivers, not only as advertised, but also according to your localization strategy’s specific needs. Your tool evaluation criteria may depend on where localization takes place in your development cycle, if you’re a freelance localizer or selecting enterprise localization solutions, and how well the tool manages your (or your client’s) language assets.

First and foremost, there are some standard features that most localization tools include that you must evaluate, such as:

  • Does it allow you to manage your translation memory files and how does it leverage them within a localization project?
  • Does it allow you to manage your glossary or term base files and how does it leverage them within a localization project?
  • Does it have project management features? Are any of those features customizable to fit individual or unique workflows?
  • Are you able to define segmentation rules for source files, target files, and translation memory files?
  • What file formats does the tool support?
  • Does the tool offer machine translation? Is it rules-based, statistics-based, or a hybrid model? How much data is require to output decent quality translations and must you provide that data? What languages does the machine translation feature support?

Beyond these, Microsoft recommends that you also add the following list of criteria to your evaluations. Microsoft says, “Besides the ability to translate the string resources within the localization tool, a few other tasks the tool should perform are to:

  • Change the names, sizes, and styles of the UI fonts.
  • Resize, move, and hide controls as necessary.
  • Change the encoding of the text.
  • Replace graphics and icons with localized ones.
  • Modify keyboard shortcuts.”

Do you use any additional criteria to evaluate localization tools? If so, please share below.