When I was a kid my parents were considering a Dunkin Donuts franchise in a spot near our house. They were in a position to pony up the franchise fee and with me being 15 or 16, they were thinking they could save on some labor costs. At the time the profits for a busy franchise were north of $300k a year, so I was thinking new car, bigger house, nice non-working college career... :)
That didn't work out, but I enjoyed many a Dunkin Donuts cup of coffee and a muffin before class in grad school. I know it wasn't that good for me, but after waking up at 5 to go row for the college rowing team for a few weeks, stopping by there was a good way to start the day.
It got me thinking when I saw this article on Dunkin Donuts v Starbucks and how Dunkin Donuts is using business intelligence systems to try and ensure their franchise deals move quickly to allow them to grow rapidly and compete with Starbucks, who doesn't use franchises.
It's an interested way to ensure they are better organized and things don't fall through the cracks, which can happen when you might have multiple people working on a deal with potential franchisees. By looking at deviations from the norm for various sections of the deals and then allowing a drill down, this is a great use of technology.
But I'm not sure this requires hard code MDX and cubes. More I think you need a data warehouse and a series of jobs running that can perform basic analysis on your workflow or CRM data. Now that is definitely the type of BI system that I think can get implemented in a more widespread fashion.
If you can make the time for your IT guys to think of how to implement it.
Ralph Kimbal thinks because it is not standardized like the Algebra end, it can get convoluted, so a few good men and women who can think calculus wanted to create the OLAP standardizing committee. Microsoft should fund it and call it the Jim Gray OLAP standardizing committee, he helped to standardize the Algebra end so the calculus end committee should be named for him.