An interesting article here on the pricing scheme for "Office Business Scorecard Manager 2005" which lets out a few more details on the new Excel server functionality:
Despite the fact that Excel 12 will have much better AS integration (the Excel blog promises more details ‘in a few weeks’ in this recent posting, so I’m keeping my eyes peeled), as I and several other people have remarked, an Excel server will end up treading on the toes of AS in a few scenarios. Smaller apps, yes, when all the data can be put in the spreadsheet (and you can now put more on a sheet, as this post on the Excel blog notes) and also financial apps where you are most likely to have a group of power users who prefer to build things themselves without IT involvement. Take, for instance, this comment about the Excel server from the Information Week article:
"The main goal here is to ensure workgroups are seeing and working with the same numbers"
Which is, let’s face it, another variation on the theme of ‘one version of the truth’ that BI vendors have been playing for years.
In the short/medium term having an Excel server is a development to be welcomed but in the long-term is there going to be pressure to improve the scaleability/performance of the Excel server so it can handle larger and larger apps? (Isn’t this how Essbase started out as a product?) Microsoft are going to have to be careful to position AS and Excel server appropriately. And also it strikes me that another undesireable effect that this will have is that users will want to do more and more of their calculations in Excel, where they can define them themselves, and only take the raw figures from AS – so that formulae end up being replicated all over the place with many subtle variations creeping in, taking us away from that ‘one version of the truth’ goal.
