You should start slow and have a definite scope of work that you need to get accomplished. Pick one thing that you need to get done and implement it. Start with that, and then you can expand from there instead of trying to drink from the fire hose and get all that information brought in at the same. You need to try to limit your scope.

I think an overall architectural understating is required. Also, you don't want to play around with multiple technologies and multiple vendors. I suggest going with one vendor who can cover the whole technology landscape.

My advice to others would be that, if they'd manage to sort out the indexing problem, it's a wonderful product to use. The biggest lesson I've learned is the deviation of the time. On a scale of one to ten, my rating would be an eight.