This is Part I in a two-part blog from Uwe Muegge, Senior Director at CSOFT International and Coordinator of the MA program in Translation and Localization Management at the Monterey Institute of International Studies. This article was originally published by the Globalization and Localization Association, the world’s largest non-profit trade association for the language industry. For more information, visit www.gala-global.org.
Understanding that translated terms are as important as source-language terms
The names for the products/services an organization offers and the features/functions of those offerings typically make up the bulk of an organization’s termbase. While these items are certainly linguistic assets, they are primarily thought of as core intellectual property, which is why the creation of these terms typically involves a team of experts from multiple domains such as product management, engineering, technical communication, marketing communication, legal, etc.
However, once a term is available in the source language and the need for translation arises, more often than not, the only expert involved in the translation of terminology is the translator. In today’s business environment, most translators work as freelancers, who typically work for multiple clients/agencies and by necessity have only a limited understanding of a specific product, its competitive situation, etc. This is especially true for new products.
After investing a lot of time and effort in developing terminology in the source language, when organizations completely outsource the terminology translation process, they are essentially leaving the development of target-language terminology to a (typically anonymous) translator, i.e. someone whose credentials and level of subject-matter expertise may not have been thoroughly assessed by the translation buyer. This practice amounts to letting the key words that in many cases have a major influence on the marketability – and usability – of a product or service be determined by a person who is typically neither close to the client company, let alone the product.
Having translated terms validated by a subject-matter expert, e.g., an in-country technical or marketing manager (either internal or with a distributor), helps close this gap and ensure that translated terms are in fact suitable and appropriate for the target market.
Enabling translators to translate with confidence
Many translators are perfectionists: For them, good enough is simply not good enough! This type of translator will question the validity of glossary terms – regardless of whether or not those terms were provided by an agency or a fellow translator. Speaking from personal experience as a longtime professional translator, I can say that the terms I was most suspicious of, were the ones I researched/created myself, being fully aware of my limitations. The result: Inconsistent use of glossary terms due to the fact that the translator discovered ‘better’ terms in the process of translation, or an endless number of comments and annotations that explain why a glossary term is not the best choice in given context.
A glossary that bears the seal of approval by the client or a designated representative, on the other hand, should convince every translator that there is no need for discussion or exploration and that the terms in such a validated glossary can be used safely.
Shortening or even eliminating in-country review
In-country review of translated documents is currently considered a best practice. The purpose of this quality assurance step is to make sure that a given translation meets the requirements of its target market. In a scenario where the translation service provider has implemented a comprehensive QA program (using ASTM F2575, SAE J2450, or another translation quality standard), incorrect or inconsistent use of terminology is one of the most common error types typically found by in-country reviewers. And it’s not a particularly surprising occurrence, considering that the terminology used in the preparation of a translation has not been validated, and thus authorized, by the client.
If an in-country subject-matter expert validates terminology before translation, reviewers no longer need to check terminology (terminology verification can actually be performed by automated tools), which should dramatically reduce the time required for in-country review. Some organizations with mature translation (and terminology validation) processes find that for many document types in-country review is no longer necessary: The combination of long-term relationships between translation buyer and service provider, detailed instructions for translators, availability of language-specific style guides, and, of course, validated terminology, reliably produces translated documents that do not benefit from an additional QA step.
Avoiding expensive and time-consuming changes late in the process
It is common for in-country reviewers to insist on reviewing the translated text – and the terminology included in it – after all other work such as translation, editing, and DTP is complete, i.e., right before publication. The reason for wanting to view a translated document in its final form is, of course, that in-country reviewers wish to check completeness, text placement, formatting, in addition to the correct and consistent use of terminology.
In-country review is typically the stage where many, if not most terminology issues arise. However, requesting terminology changes (or any other text change for that matter) just prior to publication means that not only the translated text, but also the document in its native publishing format (e.g., InDesign, FrameMaker, Flash, etc.) will have to be adjusted. The fact that this type of change requires the involvement of DTP specialists has two serious consequences: 1) Terminology changes late in the process are typically much more expensive than textual changes that can typically be performed by translators. 2) Making changes late in the process requires more time because DTP specialists often are not familiar with the target language, and the changes they implement typically need to be re-reviewed by a translator.
Many commercial translation projects are tied to the launch of a product or service on the global marketplace. In fact, translation is in the critical path of most of these launches, which means that a given product or service can only be introduced internationally once translation is complete. By the same token, any efficiencies gained by streamlining translation (for example, by reducing the time for in-country review) enable the organization to introduce the product or service more quickly.
The fact that a given product can generate revenue potentially weeks earlier just because a termbase had been validated before translation began should be a powerful motivator for implementing this additional step in the localization workflow.
Minimizing overall translation cost
Yes, adding validation to the terminology development process does cost money: You will have to identify and train an internal or external subject-matter expert (ideally the same person who reviews translations in a given market), develop and document standard operating procedures (SOPs), modify your terminology management technology infrastructure, and, of course, compensate the validators for their efforts.
A number of studies has shown that terminology work typically pays for itself within the first two years (see, for example, Böcker, Gust), and this fact holds particularly true if validation is part of the terminology development cycle. Making client-validated terminology available to the translation service provider before translation begins enables them to reduce or eliminate in-country review and possibly other expensive QA steps, which would otherwise be necessary to ensure that the final translation product reflects the client’s preferred terminology.
Reducing friction between LSP and translation buyer
Any time a reviewer on the client side identifies terminology errors in a translation, the translation service provider’s qualification, processes, and diligence are called into question. This is particularly true if the LSP is employing a quality metric like SAE J2450 that rates terminology errors as the worst type of error a translator can make.
In my experience, many so-called terminology errors identified by the client are simply instances of that client not communicating terminological preferences to the service provider at the beginning of a project. Is it an error to call a ‘Flash drive’ a ‘USB stick’? It is certainly a problem if the client’s reviewer expects to see ‘Flash drive’ and the translation provider delivers ‘USB stick’.
In these situations, heated e-mails and phone calls between the two parties are as predictable as they are unnecessary. Taking the extra step of not only managing terminology on the service provider’s side but of having those terms validated by the client before translation is a highly effective strategy for avoiding those endless and fruitless discussions about perceived terminology errors.
Automating the validation process
One of the major issues with terminology validation – as with any kind of client review – is the fact that it’s an external process (from the service provider’s point of view), and as such is typically difficult to manage. Even if the LSP is using a translation management system, more often than not, routing glossaries for validation require manual intervention by the project manager.
However, a dedicated, server-based terminology management system may offer features for automatic notification of assigned terminology validators, progress tracking, and automatic creation of an audit trail.
One of the software tools that enable enterprises to automate large parts of the terminology validation process is TermWiki, a web-based tool for collaborative terminology management developed by CSOFT.
Giving structure to a process that occurs anyway
If (in-country) client review is part of your translation workflow, you already have a terminology validation process in place. But correcting terminology problems late in the translation cycle is inefficient and expensive. Besides, chances are that at the final review stage, when many people operate in panic mode (OMG, we are already late and over budget!), chances are that the terminology changes the reviewer requires are implemented in the document under review but not updated in the glossary.
The smart way of validating terminology is moving that part from the end to the beginning of the translation cycle and making sure that all terminology changes are captured and reflected in the corporate termbase. Giving translators a client-approved, up-to-date, project-specific termbase ensures that translators use the right term the first time.
Ensuring client satisfaction
My definition of terminology, which is markedly different from the relevant ISO standards such as ISO 704 and ISO 1087, is this: terminology is the set of terms that the client cares about a lot. Following that definition, using the ‘wrong’ terms in a translation is a major quality issue.
To make sure that the client sees their preferred terms in a translation, it is not enough to just manage terminology on the LSP side. Without terminology validation early in the translation cycle, translators (and editors!) on the client side may be using the ‘wrong’ terms consistently all the way until in-country review, at which point terminology changes come with a hefty price tag.
Instead of correcting quality into a translated document late in the translation/product launch process, translation buyers and vendors should agree on proactively validating terminology before translation if at all possible. With validated glossaries in place, translators will use the ‘right’ terms (the ones preferred by the client) the first time, ensuring client satisfaction without the need for an emotionally charged, time-consuming, and expensive correction process.
References and links
- ASTM. F2575 – 06 Standard Guide for Quality Assurance in Translation[online]. West Conshohocken: American Society for Testing of Materials, 2006 [cited 7 June 2011].
- Böcker, Martin. Die Investition in Terminologiemanagement: eine Wirtschaftlichkeitsbetrachtung für den Maschinenbau [online]. Stuttgart: tekom, 2010 [cited 7 June 2011].
- Gust, Dieter. Terminologie – (k)eine Kostenfrage? Wirtschaftlichkeit von Terminologiearbeit. Produkt Global [online]. 2007, no. 6 [cited 7 June 2011], pp. 32-34.
- ISO 704:2009 Terminology Work – Principles and Methods [online]. Genève: International Organization for Standardization, 2009 [cited 7 June 2011].
- ISO 1087-1:2000 Terminology Work. Vocabulary. Part 1: Theory and Application [online]. Genève: International Organization for Standardization, 2000 [cited 7 June 2011].
- Muegge, Uwe. TermWiki: Terminology Management Just Got Easier. tcworld [online]. 2011, March 2011 [cited 7 June 2011].
- SAE. J2450 Translation Quality Metric [online]. Warrendale: Society of Automotive Engineers, 2005 [cited 7 June 2011].
- SDL. Terminology Survey 2010 [online]. Maidenhead: SDL, 2010 [cited 7 June 2011].