We’ve written extensively about translation quality from the buyer’s perspective, arguing that companies purchasing translation services should assume a more proactive role in defining their requirements, developing their own metrics, and communicating with suppliers about how they intend to evaluate their performance. Last Friday, we convened approximately 30 delegates from the industry’s largest buyer organizations for a Common Sense Advisory colloquium hosted by Google in Mountain View, California.

In our earlier research on buyer-defined translation quality (”Buyer-Defined Translation Quality,” Aug08), we revealed the findings of interviews with 28 companies that translate hundreds of millions of words into multiple languages each year, most of which spend in excess of US$1 million on translation services annually. The report answered the question, “What does translation quality mean?” from the perspective of large-scale buyers of translation services. We identified the most common challenges associated with quality-related issues, and we mentioned an important finding — that only a very small number of buyers were formally measuring translation quality.

Prior to the colloquium, we asked attendees to fill out a pre-event survey on translation quality measurement — 20 individuals responded. We spotted a few trends worth noting among this group:

• Few buyers formally measure translation quality. When we asked attendees if they measured quality in a consistent, documented fashion, only a slim minority answered “yes” to this question.
• Buyers struggle with consistency in measurement.  In response to our question about the top challenges with regard to measurement, consistency was the most popular answer, followed by lack of time to devote to measurement.
• Most buyers are new to quality measurement. When we questioned buyers on how long they had been measuring quality, the majority had been doing so for less than a year.
• Self-developed processes reign supreme. We asked buyers how they developed their quality measurement processes. The majority reported home-grown processes, while the second largest group said they used a hybrid of their own processes and the LISA QA model. Tellingly, not a single buyer said that they used a process recommended by their language service provider.

This last point highlights the fact that buyers are becoming more sophisticated and want to take greater control of quality measurement. It is not something they feel comfortable leaving solely in the hands of their language service providers. Why not? Globalization leaders from Dell (Wayne Bourland), Google (Samir Patel), and Motorola (Elsebeth Flarup) presented case studies on how they evaluate the quality of their vendors’ work, followed by group discussions and attendee presentations of best-case and real-world scenarios for translation quality measurement.

Our latest members-only Quick Take, “Buyers Step Up Their Translation Quality Measurement Efforts,” reviews the major findings from the colloquium, as well as Common Sense Advisory’s views on what needs to happen to move translation quality measurement to the next frontier.