Translating Help Desk and Knowledge Base Documentation
Customer support for technical products and services is a big enough nightmare in your native language, i.e. the language the product or service is developed in. Adding multiple languages can multiply the headaches. However, if your tech content team plans their processes with the understanding that multiple languages will be needed, you can simplify the process.
Localizing your customer support tools in real-time (or close to it)
With the use of faster moving agile-style design and development processes, the number of changes being made to the product typically means much more frequent product releases or upgrades, sometimes as often as weekly. When you add languages these changes need to be supported in those languages. These rapid changes and the inherent complexity they introduce often drive the decision to move to a more sophisticated content development strategy. This article is an overview of how these systems work and how to align your translations with your product development timelines in real-time or at least close to real-time.
Paralleling ‘agile’ development processes
Agile design and development processes are becoming the standard process for many technical products like software and software embedded in hardware. These processes break development tasks down into small pieces that can be created, tested, and released on a more or less constant pace. This is in contrast to the waterfall approach where many elements are developed over a longer period of time and released as a ‘major’ upgrade.
The challenge with translation of these rapid releases is keeping them timely and in sync with their release dates. In a worst case scenario, translations lag behind and product releases are either held up or only released to certain languages, a process that can get unwieldy at best and damaging at worst.
Fortunately agile processes are managed with a systematic approach that uses project management tools like JIRA and, with the right workflows, your translation vendor should be able to use the same software to mirror your release schedule. There are tools designed to support these workflows that connect directly from your technical documentation teams to the translation team.
Maintaining formatting context with Component Content Management Systems (CCMS)
Component Content Management Systems are the next evolution of creation tools for technical content development. They are particularly valuable for documentation requiring translation. In fact, companies seeking an ROI justification for the costs associated with acquiring and implementing a CCMS, often find translation savings alone justify the expense and effort. Here’s why:
Content is stored in a central database that writers, editors and project managers access. There are no additional versions floating around.
The systems are XML-based, meaning their file format is readable by all major translation memory/ management tools
Most CCMSs can send files directly into translation memory software, typically the first step in the translation process
The translator does the translation, the translation memory stores the translated content for reuse, the content is reviewed and edited, and then returned to the CCMS for publishing.
There are no version control issues, which cuts costs due to errors and redundant effort.
The content in the CCMS is structured with meta tags, enabling reuse of individual ‘chunks’ of content based on their type (see next section)
The CCMS is capable of formatting the content for a variety of media, on the fly, enabling almost instant digital publishing to the web, knowledge bases, print, etc.
Added together, these functions ultimately give you the ability to easily publish your translated technical content to specialized formats like Help Desk and Knowledge Base software. This does require development of template tools for those specialized formats but once they are in place you can use them over and over again.
The sauce that makes this work: understanding technical content structure
In technical documentation, writers primarily create three types of content:
Concepts. Blocks of text used to explain things
Task lists. Ordered lists of how to do things including operation instructions, installation instructions and and maintenance procedures
Reference. Specifications, legal boilerplate, parts lists, etc.
Even if the documentation being written in MS Word, it generally can be broken into these distinct types. The advantage of the CCMS is that each of these chunks of content is technically its own discrete file with its own set of metadata that ‘tells’ the system how to display and find the content. To assemble a document like a manual, you assemble these different components in the desired order using tree-like maps, not unlike tables of contents. From a translation perspective, this ability means you do not pay for content, like legal disclaimers, that appears in multiple documents. It is translated once and then stored in the CCMS for reuse. Any changes made to it change every instance of it in use across various digital media.
Eliminating duplication of effort with translation memory software
On top of this protection from unnecessary translation, the CCMS ‘pushes’ each content chunk into the translation provider’s translation memory which takes this one step further, identifying snippets of text that appear more than once, so they only need to be translated one time. And once this information is stored in the memory, it will be flagged as having already been translated previously. In the case of the aforementioned legal disclaimer, for example, the memory would simply compare the text and if it is a match it will not be sent to the translator a second time. Because translators typically get paid for by the word, this can result in significant savings.
In a perfect world…
As a language service provider, we’d love to see all of our documentation clients using a system like a CCMS, especially for technical, legal, and life sciences content where accuracy is critical and there may be many versions for different countries. Unfortunately, we are often still working with large, unwieldy documents without structure. A lot of time is spent understanding the context of these files and each requires formatting of the completed translations. For instance, in a lengthy PPT, text would be extracted, translated and reviewed, and then rebuilt in the PPT template for each language. If it were being sent to a CCMS, the system would manage this publishing process, giving our clients much better control while saving significant time and money. We can always dream…