Hello,

I'm in charge of sending out translation requests to a translation company so, as sort of a project manager, I'm prepping the resources and source file to make it both easy for the company/translator, and cost-efficient for my company. Please tell me if I'm doing anything wrong or could do anything better.

I first took the source file (Japanese ppt) and ran it through Trados and created a project with a default TM to see how it would segment. It looked awful so I wanted to adjust the segmentation rules so that it would segment properly. I was having trouble achieving this and no one could help.  https://community.sdl.com/solutions/language/translationproductivity/f/90/p/13299/46901#46901

So I just split and merged the segments as I liked.

Next I thought it would make sense wanted to create a TB for the project. Do translation companies do this naturally? Is this even necessary? 

Then I am curious about whether or not I should leave merge segment on or off or what sort of settings I should have for the translator. Does anybody whos handled lots of packages tell me what they would ideally want?

Also is it necessary or preferred to perform a pre-translate from my other TMs? Would the company do that anyway before starting?

Should I analyze the file and get word count/fuzzy match count and whatever necessary data before asking for a quote? Or should this data be used the client-side while negotation?

Sorry, lots of questions. But any insight anyone could offer would be appreciated.

Anyone from the client-side has any advice as well for their standard process, that'd be fantastic.

Best regards,
Keenan