When handling an update to a previously translated project, it may seem like it would be easier to simply update the original file directly based on the new English version. However, this approach raises the question of how the updated content gets into the translation memory. If we complete the update directly in the original files, there would be no translation memory entries reflecting the update.
We suggest an alternative approach. The update can be handled in the source files using our translation management system. The process is simple. We import the project into our system where each sentence is checked against the translation memory database. Any previously translated sentences are committed to the project and discounted from the total cost. The only items considered for the new project would be the modified or additional content.
By following this approach, the next update will properly leverage the translation memory database insuring consistency across projects, reducing the time to completion and securing a lower total cost for translation.
We are happy to work with either approach but it is important to understand the risks and benefits of each option.
For additional information on how translation memory works please see this blog post.