What is a Quality Manager in a translation company doing? Checking the quality of the translations, of course. That’s what one would think immediately. But just like many other jobs, the job of the Quality Manager is changing, and just like with other jobs the reason is: data. Here are five reasons why modern Quality Managers should be obsessed about data.
Volumes of translation work keep growing, and it is simply impossible for a Quality Manager to proofread and correct all translated texts. To scale up the Quality Manager must work methodically, set up proper processes and perhaps rely on sample inspections of quality. Deciding on samples requires a statistical approach. What represents a good sample depends on analysis of data. Which data points exactly need to be taken into account may be different for different content profiles and different vendors and translators.
Gone are the days that one quality level would fit all content, all audiences and all purposes. Today customers can ask for high quality translations or good enough translations. How on earth can a Quality Manager make the distinction if s/he does not have any data to fall back on? Definitions are not enough. Penalties and weighing scales linked to error types will bring some structure into the process.
Quality assurance or management can cost 15% to 25% of the total translation costs. A lot of time and money is lost on disputes over quality. In case of doubt, a second editor may be asked to check the translation or an in-country reviewer or subject matter expert is hired to do an inspection. The more opinions the more confusing the discussion on translation quality usually gets. In the worst case this may lead delayed product availability and loss of opportunity and revenue. The modern Quality Manager simply cannot do without an objective quality metric, which is filled with data.
If you can’t measure, you cannot improve. That’s a simple and true statement. Quality Managers are responsible for comparing the quality of vendors and translators, and for resolving quality issues when they occur. But how can they do that if no objective data are collected and aggregated. The modern Quality Manager sits in front of a dashboard where s/he can track and compare the quality of translation projects, vendors and technologies.
All of the previous four arguments are good for the tracking the past performance, but what really matters is of course how we learn from the past to build a better future. The future-proof Quality Manager in many ways becomes a data analyst who uses data to predict quality. Aggregated data on translation quality and productivity help us to develop algorithms that assist in resource allocation and selection of MT engines and processes that will give us the highest confidence that the targeted quality will be achieved.
Jobs in the translation industry are changing. Just like in many other industries data become increasingly important. The modern Quality Manager becomes a kind of data analyst in his own right. TAUS has developed the Dynamic Quality Framework and the Quality Dashboard to support Quality Managers taking a more methodical approach to quality management.