Microsoft introduces Power BI in the cloud

Fredrik Tunvall, Analyst, Software – Information Management, Tony Baer, Principal Analyst, Software – Information Management

Microsoft is betting heavily on the cloud for delivering self-service BI capabilities. Power BI builds on existing Microsoft Excel building blocks, including Power Query and Power View, to provide a single self-service client based on the familiar Excel spreadsheet as the user interface. It is a welcome simplification of Microsoft’s often confusing BI product stack that includes overlapping capabilities. The Power BI for Office 365 is priced competitively versus Tableau Online, the closest comparable product.

Making BI native to Excel

With many millions of users, Microsoft Excel is arguably the world’s most popular BI tool. With Power BI Microsoft sharpens its focus on the BI end user by leveraging the ubiquity of Excel, allowing users to load data, manage queries, visualize the answers, and manage connections to a variety of back-end data sources inside the familiar spreadsheet UI. However, Microsoft’s BI portfolio has been a complex patchwork of frequently overlapping tools, applications, and add-ons from SQL Server, SharePoint, and Excel.

Microsoft intends to make BI functionality native to Excel. However, in the current Excel 2013 release, BI functionality is still a mix of add-ins organized under the “data” tab. But with Power BI, Microsoft is starting to simplify the onramp to analytics, using cloud-based Excel as the launch platform. However, Ovum is concerned that Power BI requires customers to upgrade to Excel 2013 Premium Plus or Office 365. That might involve a far more comprehensive upgrade compared to installing a BI point solution. Consequently, it may take some time before we see how successful Power BI will be as it is tied to an overall Office upgrade.

With Power BI Microsoft bets on the cloud

With its new Power BI for Office 365 release, Microsoft is extending the self-service query, reporting, and visualization functionality of its on-premise release to the cloud, and is aggressively pricing it to tap as much of the Excel user base as possible.

Power BI for Office 365 sits atop the cloud-based SharePoint service. Prior to this release, SharePoint users on Office 365 could access Excel files for query only up to 10MB in size; with the new Power BI service, that threshold goes up to 250MB. This is important, since Power BI does not support direct connections to databases, and data instead needs to be stored in Power Pivot models before it can be queried and visualized.

Leveraging SharePoint, Power BI offers the ability to share queries, and the ETL routines that are used to populate data for them. Queries are managed through a metadata catalog that allows other users to leverage and reuse existing queries.

The new cloud offering also leverages the Windows Azure data platform, providing access not only to SQL Server, but also the HDInsight service that exposes data from Hadoop. It leverages Excel’s existing connectors to back-end data sources such as SQL databases and Microsoft Exchange, and Windows Azure’s existing database gateways that allow live, bidirectional exchange and update of data with on-premises sources.

A unique feature of the cloud-enabled Power BI is a new capability to search and connect to publicly available APIs for “Open Data” sources on the Internet, enabling users to search and connect to publicly available data sources. This feature is a logical companion to Power BI’s query function by enabling administrators to curate a library of data sources that can easily be accessed by BI users. However, Ovum believes that for the feature to be effective, enterprises need to have data quality measures in place to ensure users only connect to proven and trusted sources.

In Power BI for Office 365 there is also a natural-language querying feature, named Q&A, which offers a Google- or Bing-like search to find answers in the data. Similar to IBM’s Project Neo and Oracle Endeca, Q&A targets nontechnical users unfamiliar with SQL. However, Q&A only works with Power Pivot data models; unstructured or semi-structured data sources are not currently supported.

Mobile strategy aimed at tablets

Power BI is targeting tablets, not smartphones. On initial release, Power BI will support Windows 8 Surface devices, and will add iOS support for the iPad at a later date. But lack of native support for Android is a real omission that Microsoft should redress, given the growing popularity of Droid-based tablets.

Ironically, as Microsoft does not view smartphones as appropriate targets, it is not planning to develop a version of Power BI for Windows Phone. Nonetheless, like Tableau, MicroStrategy, and Qlik (formerly known as QlikTech), it is increasing its support for an HTML5 universal mobile client.

Licensing optimized for existing Office 365 users

Microsoft is pricing competitively versus Tableau, which also offers a cloud-based self-service BI service. The annual cost per user is $240 (promotional price) if you already have an Office 365 subscription (which packages the necessary SharePoint and Excel components utilized by Power BI). A purchase of Power BI and an Office 365 subscription costs $480/user annually compared to $500 for Tableau client online. But there is the need to create content for those self-service views. The Microsoft clients are included in the Office 365 pricing, versus $1,999 for Tableau (for an on-premises copy of Tableau Desktop Professional Edition).

So that makes the Microsoft solution appear far cheaper; however, the comparison is apples to oranges. Power BI is an extension of the general-purpose spreadsheet Excel, while Tableau offers a more robust solution for the data visualization and discovery subset of BI. In particular, Tableau’s visualization is far more comprehensive than Power View (which accompanies Power BI), while Microsoft’s Power Query (also bundled with Power BI) has much deeper data transformation and cleansing capabilities compared with what Tableau offers.

So, on one level, the decision pitting Tableau against Microsoft for the cloud-based self-service is a classic scenario of rich (Tableau’s more vivid visualization capabilities) versus reach (the familiarly of the Excel spreadsheet and comprehensive self-service ETL capabilities).

Further reading

Business Intelligence and Analytics Fundamentals, IT014-002869 (January 2014)

2014 Trends to Watch: Business Intelligence, IT014-002810 (October 2013)

Related Stories

Leave a comment


This will only be used to quickly provide signup information and will not allow us to post to your account or appear on your timeline.