Microsoft BI Review 21 of 55
Microsoft BI Review: "Think hard on your BI goals" BIUnspecified7.9615101
Haibo Yang profile photo
September 26, 2016

Microsoft BI Review: "Think hard on your BI goals"

Score 10 out of 101
Vetted Review
Verified User
Review Source

Overall Satisfaction with Microsoft BI

We have been using a Microsoft BI stack (SSIS SSRS and SSAS) to support our SaaS solution for almost a decade. Over sixty client organisations are now using our SaaS solution to manage their operational risks and related activities. With Microsoft BI, clients can use Report Builder to access our data sources to design, develop, publish and distribute their own reports. On the other hand, we as a service provider can use Visual Studio Data Tool to create and maintain data sources/models. I mostly get called in to design and develop sophisticated dashboards and smart KPI reports (often programming-intensive in nature) for clients when the complexity goes beyond dragging and dropping plus simple VBA expressions. Overall, our clients are happy with the reporting flexibility delivered by Microsoft BI and we ourselves are very positive with the future of this product.
  • Point 1. User defined automation of report execution and distribution. Microsoft SSRS so far is one of the most user friendly report scheduling and distribution platforms available. Our client users, often non-technical business people, can subscribe to any reports they have access to on the report server and make a customised execution by setting up parameter values, export formats, receipients, etc. etc. Many users use this feature to monitor their action lists and risk profiles on a regular basis. They absolutely love it!
  • Point 2. Extensive programmability. Programmability has always been a great strength of many Microsoft products. Adding to my point 1, take Microsoft SSRS for example, it comes with a great deal of programmability. This means what client users need do in point 1 to set up the report execution and distribution by themselves, can now be programmed and completed automatically. One trick we often do is to program on SSRS for automatically executing and distributing a report using different parameter values to generate different results and then send to the email boxes of tens, if not hundreds of line managers within client organisations. Every line manager will only receive the results relevant to his/her own business unit(s). Once set up, a client organisation can save hundreds hours of work on Excel spreadsheets each month. Clients are willing to pay you a fortune for such a level of automation in reporting process!
  • Point 3. Flexible integration with SSAS. Instead of praising the more techinical features such as partitions and actions shipped with SSAS, I'd make my point 3 to be more business user friendly by emphasising the integration options of SSAS. Excel, Power BI, SharePoint, and third party tools such as Tableau, can all be easily and nicely integrated with SSAS objects. Not to mention since MSSQL 2012 you also have got the choice between Tabular models and Multidimensional models. Your business analysts will love the flexibility SSAS can provide!
  • Point 4. Stability of the IDE. Of course this one is to me when I'm in a BI developer mode. Using Visual Studio to develop SSAS, SSIS, and SSRS objects is a relaxing experience and will be good to the longevity of your developers. Why? Because VS is stable enough to not crash your developers' computers. As a matter of fact, VS has never crashed my OS since 2008...I mean it.. though a few times before 2008...Unlike some other seemingly simpler IDEs which may freeze or overpower your OS while processing your design changes... VS is a powerful yet stable tool and your developers will love it.
  • Report Builder 3.0 shipped with MSSQL 2012 is a nice free tool but our client users sometimes encounter problems such as the tool automatically shut down without saving the changes being made. We haven't implemented the new RB with MSSQL 2016 yet thus cannot comment on the latest version.
  • As a SaaS provider we see being able to provide self-service BI to our client users as a competitive advantage. In fact the MSSQL enabled BI is a contributing factor to many winning RFPs we have done for prospective client organisations.
  • However MSSQL BI requires extensive knowledge and skills to design and develop data warehouses & data models as a foundation to support business analysts and users to interrogate data effectively and efficiently. Often times we find having strong in-house MSSQL expertise is a bless.
I'd rather not say to avoid unnecessary arguments. Thanks.
Most suitable scenarios:
-Large scale report automation and distribution.
-Self service BI for internal and external users.
-Relational databases and multidimensional models.
-Comprehensive security & access control.

Less appropriate scenarios:
-Non-relational databases
-Low budget
-Tight timeframe

I'd invite anyone reading this far to think hard on his/her goals with BI. Are you trying to build a solid and endurable BI service for your clients or your own organisation? Or do you just need to have some quick visualisation of the data you have to make strategic or operational decisions in a few weeks time?

Implementing a MicrosoftS BI stack takes time, knowledge, and skills, none of these comes cheaply these days. If your answer to my first question is "yes", go ahead and study Microsoft BI a bit more then make your decision on your own. If you see my second question is most relevant to you, go and grab a web-based BI tool such as SiSense, Tableau, Splunk, and so on. Take the free trial option and see if you can test your ideas fast and at a lower cost. Good luck!

Microsoft BI Feature Ratings

Pixel Perfect reports
Customizable dashboards
Report Formatting Templates
Drill-down analysis
Formatting capabilities
Integration with R or other statistical packages
Report sharing and collaboration
Publish to Web
Publish to PDF
Report Versioning
Report Delivery Scheduling
Pre-built visualization formats (heatmaps, scatter plots etc.)
Location Analytics / Geographic Visualization
Predictive Analytics
Multi-User Support (named login)
Role-Based Security Model
Multiple Access Permission Levels (Create, Read, Delete)
Responsive Design for Web Access
Not Rated
Dedicated iOS Application
Not Rated
Dedicated Android Application
Not Rated
Dashboard / Report / Visualization Interactivity on Mobile
Not Rated