I was preparing a quotation for a client today and the question came up – Why aren’t the number of 100% translation memory match deductions higher? I was sure we didn’t make many changes to the content.

Per the clients request I decided to review the files and scroll through the segments and sentences to see where we were losing matches. I found two interesting examples that certainly caused some of the issues.

In this example the original sentence was
A Customer Service Representative will make every attempt to resolve the issue over the phone.

the new sentence is
Customer service will make every attempt to resolve the issue over the phone.

This change may have only affected a few segments.

In our second example a change was made that affected many segments across the entire manual.

In this example the original sentence was
When you encounter a difficulty, please use this guide first to answer your questions and resolve the issue.

The new sentence is
When you encounter a difficulty, please use this booklet first to answer your questions and resolve the issue.

Again another very subtle change guide was changed to booklet. However this type of change could affect many segments. Globally changing a term that is used over and over again in a document will eliminate many 100% matches to and demote those segments to Fuzzy matches.

This additional cost could be significant. Especially if you multiply that cost by the number of languages in your project. A fair amount of our customers publish in 28 languages or more. Even a small $100 increase per document is material when you multiply by 28!

Writers should carefully weigh the importance of changes and fully understand how many segments the changes will affect.

Share This