Microsoft BI vs SAP BusinessObjects Business Intelligence Platform comparison

All the traditional BI platforms including Business Objects and Microsoft Reporting services and Analysis services require IT involvement almost at every step in preparing the data and report.

Self serve BI is the promise to these business analysts without technology background. However following characteristics are a must to meet the self serve BI dream.
- BI tool should be capable of reading data from its source without a dependency on ETL or a warehouse.
- While a dimensional model gives most flexibility for ad hoc data analysis, it brings a overhead of consistent modeling mindset requiring very technical background.
- Ability to convert grid data into visualization and vice versa with few clicks
- Ability to mashup multiple analysis from multiple sources on to a single screen.
- Finally a framework that let's end users seamlessly build their analysis while IT can throttle, govern, audit and scale end user data needs with a great amount of automation behind the scenes as a continuous process as opposed to be a pre process.

Two such platforms I have come across are
1) Tableau
2) CarbonBI

These solutions seem good for Visualizations. I like Pentaho personally. Wondering why the this suggestion hasn't been made??

Sap business objects can provide a sophisticated self service solution that is very easy for the end users to engage with for both ad hoc analysis and report writing and distribution. However as with all Bi solutions the back end data warehouse must be designed intelligently and business objects universes configured correctly. The same thing really applies no matter what toolset you select. If you already have business objects then it makes sense to ask IT to set it up as a self service solution rather than look for another technology. If IT do not have the skills then look for a good consultant to perform a review of your BI solution and make recommendations.

Nick,

Good comments similar to the points I was making. I think that it is still
important to consider how much data you expect to be dealing with, the
tool's analytical architecture (ROLAP or MOLAP), the sophistication of your
analyst end users, and how complex your reports are likely to be. If you
or the analysts expect that solution development is going to be in the
hands of the analyst, then the tool needs to be relatively easy to learn.
On the last point, if you expect a lot of slicing-and-dicing you need an
architecture that will support the high indexing load. Anyway, success and
use acceptance is not just a question of apparent simplicity and seemingly
low cost.

regards,

Keith Breedlove
Polyglot Analytics, LLC
Groveland, FL

I suggest Power Data, the new Microsoft develop.

Try Tableau.

I would suggest looking at Tableau for requirements of self-service nature. The success factor for a self service tool depends on the ease-of-use for the end-user who is less proficient in IT skills and the range of tasks it allows the end user to accomplish. Tableau scores highly on both these parameters. Backed by a well designed data mart, Tableau can be the solution that pretty much allows the end user to replace the need for IT. It has excellent training materials available in one-click and many forums where people are ready share their cool experiences. Developing a report in Tableau for me was more like playing a video game, a throughly enjoyable experience to get to a cool end-product. You want the end-users to cherish the process of creation and Tableau does that with ease.

I would focus on Tableau and MicroStrategy (we went with MSTR several years ago to supplant BO), although QlickView has its proponents for ease of use...