Yeah, I've used it to oxidise benzophenone hydrazone to diphenyldiazomethane (the other method I used was HgO/KOH, guess which was nicer...).
Nice, clean, non-toxic reactions with very simple workup. You have to use a large excess (around 15 - 20 eq) because only the surface reacts (so the core is still magnetic CrO2 after the reaction is complete), which also makes it impossible to calculate your stoichimetric ratios - but if the reaction is clean then that's no big deal, and as you say it can be recycled.
Stirring the mixture is a bit of a pain - you either need an overhead or you have to accept the fact that you'll never use that magnetic stirrer bar for any other reaction. If you go for the magnetic stirrer approach the Magtrieve doesn't move - you just need to have the solution moving vigorously and you get sufficient flow over the surface of the solid. It also gets all over magnetic retrievers and is very difficult to remove. Also, put a sheet of paper on your stirrer plate so you can lift off any stray Magtrieve particles easily - otherwise they'll probably be stuck there forever. All minor annoyances though, it's definitely worth a shot and it's especially good for larger scale work.
As I say, in my case the choice was a no-brainer - method A involved filtering off >100g of liquid mercury, method B you just wave a magnet at it...
It would be cool to have other reagents immobilised on magnetic beads!