Some general observations on Microsoft BI 2010

For those of you who have seen The Princess Bride, I will quote Inigo Montoya by writing, “Let’s begin!” First, Microsoft’s latest BI stack will smartly continue to leverage SQL Reporting Services (SRS) for the delivery of reports. I personally did not see any significant or noteworthy improvements to SRS for 2010, but Microsoft continues to be a solid solution when it comes to generating pixel-perfect, banded reports.  This is especially true if you are generating a report that has large data tables that span multiple pages. SRS reports continue to support some drill-down functionality, but that is obviously lost if the reports are to be distributed as .pdf files. For classic report-driven information distribution, SRS reports combined with SharePoint for managing subscriptions and delivery work well.

 

Another area where the Microsoft BI stack shines is related to the ability to collaborate. What a fancy word…”collaborate”. Now I do not want to steal my own thunder from the third and final installment of my Microsoft BI 2010 stack review that is yet to be posted, but the key to the success of Microsoft’s solid collaboration is SharePoint itself. SharePoint 2010 looks to be much improved, both strategically and technically. The ability to share Excel files and work on them in conjunction with other users is a nice capability that will continue to grow in popularity as more business processes become adapted to SharePoint-enabled workflow. It seems to me that many business processes, like budgeting and forecasting, are going to be improved by leveraging SharePoint’s collaboration capabilities. For example, a user can be prompted to update their forecast data in a spreadsheet and that data can then be automatically fed back into a source database so the resulting analysis objects are updated on a dashboard. It almost seem like this is the way things are supposed to work.

 

Although data mining was added as a feature to Microsoft SQL Server 2005, it has been refined in SQL Server 2008 and retained as part of the “nice” features of the 2010 BI stack. Microsoft should really call it “data mining for dummies”, because in most cases it does not require an advanced statistics degree to use. The end user can pretty easily identify a table of data and apply the data mining tools against that data set to see which factors have the greatest correlation to the others. Sometimes you need a little help figuring out the most important factors in data set and Microsoft’s data mining services could be just the ticket. 

 

Finally, Microsoft has done a nice job making many of the BI stack components easy to use by incorporating many of their common navigation features. For example, in Excel 2010, the slicer and PowerPivot features are simply added to the “ribbon” toolbar that is now part of Office 2007 applications. For other BI components, they have been given wizards and other navigation interfaces that make them feel like, well…Microsoft tools. Their interfaces incorporate more clicks and menu options than some might like, but it should not feel uncomfortably foreign for most. It is interesting to see how their strategy for updating a dashboard object compares to other solutions. With Microsoft, most of your object formatting is done via properties displayed universally in the ribbon toolbar or in a large assortment of menu options. With some other solutions, you right-click on the object and that opens the formatting options that are specific to that particular object. I am not sure if one approach or another is truly better, they’re just different.

 

That’s it for now. My next post will be the last in my series on the SharePoint 2009 conference and will include some thoughts on SharePoint itself as a BI delivery tool.

 

Until next time…Shawn