The Neverending Story of Word Counts

Do you remember Michael Ende’s book, The Neverending Story? Or, like me, do you remember more the them song by Limahl’s?

Whenever I have to make a quotation for a client and I ask them for the source files, I enter in the fantasy world of word counts, which is full of  fantastic creatures that can make your word count as big as Falkor, the luckdragon, if not prepared correctly.

Our last adventure in the fantasy world of word counts took place quite recently.

We got a quotation request for translation of a website, and we asked the client for the source files. The client exported the website into individual xml files, and we analyzed them to get a word count. We used Trados Studio for that, and we got a word count that was a very nice starting point, but which we knew that was not real: more than 60,000 words.

The file was not prepared correctly for translation, so we had to prepare it ourselves. We needed to create a configuration file that Trados would use to know what is translatable and what is not. If you are a translator, follow these steps to learn how to do it. If you are a client, just skip to the end of the article, and you will be happy to know what the word count will be after all these steps.

When you create the Project in Trados Studio, you will reach a point where you have to select the files to translate. Before doing that, go to the File Types option:

When you click on File Types, you will see a list of all file types supported by Trados Studio. However, as I mentioned, we want to create our own file type, based on the files we are going to translate.

Just click on New and select the desired type. In our example, we are going to select XML.

Follow the instructions of the wizard and select if you want to create an XML file based on default settings or based on settings from an existing settings file. In our example, we are going to select the second option, and we will browse to select one of the translation files:

The Parser Rules dialog box will now appear. Here is where we need to select what is translatable and what is not. Just go through the list of rules and double click on each of them in order to select the status from Translatable to Not Translatable.

Once you are done, your file type will appear in the list of files types.

And when you add the files to translate to the project, they will all appear under the file type you created.

By doing this, when you analyze the files, your word count will be much real, as it will only take into account those strings that need translation.

In our example, we moved from a word count of more than 60,000 words to around 19,000 words. It is a big difference in our income, but if we had not done it, the client would have not accepted our quotation or he might have paid much too much.

Also, translating segments full of xml code that is not to be touched can be really annoying, so the translators will end up spending more time than expected.

However, even if we have sorted out this problem and managed to provide clients with an accurate word count that matches with what they need to translate, we still need to face another issue: how can we convince clients that they should send the full source files for quotation? Have you managed to get them? Tell us your strategies in the comments!

Related articles:

Why do repetitions have to be included in a text to translate?

• Don’t touch my source files!
• When the file to translate is sent “as is”
• The Localization Project. Part 3: Creating the Source