Document conversions are far from ideal -- heck, it's a crapshoot as to whether a document will convert between different versions of Office! The more Office features you use, the slimmer the odds that your document will come through to OpenOffice unscathed. Probably the strongest point in the video is about Excel macros. They are a powerful but obscure style of programming. I once wrote an entire expense reporting system in Excel that faithfully recreated a printed expense form. That's not going to convert. If your company has a bunch of Excel "apps" like that then you will be using Office far into the future.
So although every statement in the video is probably true, those are the exceptions and not the rule. By far, most users are not exploring the obscure feature spaces of Microsoft Office. Instead they are writing lightly formatted documents that are only a few pages long, static presentations filled with nothing but bullet-pointed text, or simple spreadsheets that sum a few rows and columns. If you're given the job of deciding whether an OpenOffice migration is possible, you'll need to determine whether the exceptions in your own organization are manageable.
One of the segments in the video did upset me. It was from a teacher who said "I've had students who turned in files that they've converted from OpenOffice with formatting problems that affects their grade." Look, if it's typical school paper I can't think of any reason why it should be necessary to use Office. A typical six-page typed-double-spaced document in Times New Roman should convert to any word processor, or be perfectly acceptable in Google Docs for that matter. When grades come down to document conversion issues and only Office is the answer, something is very wrong with our educational system.