Have been tasked to put together a quick write-up with respect to the Microsoft Power BI in comparison to QlikView & Tableau for a potential larger BI solution piece for a well-respected client.
Just sharing my research with everyone here and more than welcome for any feedback from all of you:
An brief overview of Microsoft Power BI
Microsoft Power BI is a collection of different Excel add-ins, SQL Server services features, SharePoint and/or the Office 365 Cloud. Although desktop Power Pivot add-ins can be easily installed in Excel 2010 and is already embedded along with Power View in Excel 2013, Microsoft BI on-premise server set ups are complicated, one week or longer installations and configurations of SharePoint, SQL Server and Reporting Services that also require Active Directory accounts and Kerberos for authentication.
(Note: Power View is both embedded in Excel 2013 but also as a stand-alone version via Reporting Services Power View reports hosted in SharePoint. Each version has a different mix of capabilities.
Great ETL & Data Cleansing/Validation Components
Microsoft Power Query is quite nice and has both basic and advanced self-service ETL capabilities including pivot/un-pivot, joining, filtering, de-duplicating, grouping, splitting, and transforms.
More advanced functionality can be achieved via Power Query M scripting. Although Power Query is in preview and a v 1.0 offering, it gets monthly enhancements and already has proven to be a great tool for personal data collection and shaping.
Not fully compatible for Tablet/Mobile devices yet
Microsoft Power View was designed in Silverlight. The Office 365 Power BI Cloud version of Power View has an HTML5 option in preview for some of the chart types. On-premise Microsoft BI Power View is only Silverlight today and not fully mobile device friendly.
Data Visualization not as appealing as Tableau
Microsoft Power View has basic bar charts, line charts, pie charts, a scatter chart and non-customizable Bing Maps that can show pie charts or circles – no thematic mapping or overlays.
Power View is lacking combination charts, dual axis charts and many other common chart types. Power View does not allow granular control of visualizations, axes, colors, labels, annotations, drill actions, conditional formatting or visualization display logic, visual banding, statistical reference lines or custom marks.
There is not an API for automatically building Power View visualizations. Only Power View scatter charts in can play over time periods. Power View does not support easily viewing underlying details or exporting the view data. You can do some simple drill drowns. There is no comparable export of Power View detail data shown on a view. To see the raw underlying detail data, you need go back to the back-end Power Pivot model.
Power View does not have features for changing data color variations, KPI icons are only circles, no chart object sizing, contouring and no customization of tool tips. Power View does have text sizing and limited chart labeling options.
Here are just some insights based upon the BI/Analytics communities practice and be categorized into the following aspects: Business Criteria, Data Visualization Criteria and Technical Criteria.
As mentioned before, it’s just another product comparison based upon people personal view over those products according to their own preferences and practices experience.