CT treats the tags as text in .docx files. I think hiding the tags in Word might work to avoid any conflicts with CT formatting tags. I would suggest making a test with the export after translating several segments.
Thanks for your suggestion Igor,
It turns out that CT treats the hidden HTML tags as tags, which is perfect!
Exported text looks perfect.
So my question is, what's the best way to deal with html tags in a docx file? Does CT treat these as tags and allow me to place them as required? Or does it treat them as text?
If the latter, my initlal plan was to simply hide them in Word so they won't be displayed in CT. That runs up against the problem of how do I then correctly place inline tags such as <b> and <strong> in CT.
Anyone come across/found a solution to this issue?
Note that I can't simply convert the files to html, as they also contain lots of junk text which I am keeping out of CT by setting them to hidden in Word.