My Microsoft Experience
I’m looking at Tableau from the perspective of someone who spent much of his 28-year data career in the Microsoft space, broadly utilizing all parts of SQL Server databases and tools from SQL Server 2000 through to SQL Server 2012 from the perspective of several different roles (developer, lead developer, DBA, database/BI architect). My Microsoft data platform experiences are pretty broad and fairly deep (depending on the component). In most recent years, I utilized Excel’s PowerPivot and evolved with it into the Power BI product that has a desktop component, Power BI Desktop, and a cloud deployment platform. Power BI is where we can build a point of reference to introduce Tableau. This is not a side by side comparison, but rather a very brief introduction to Tableau for someone with a Power BI point of reference.
Tableau is Roughly Equivalent to Power BI
Out of all the data platform topics I’ve mentioned above, Tableau resides in a similar space as the Power BI product. Clearly, Microsoft did not even compete in this space until after Power BI became its own thing and evolved from being the PowerPivot plug-in for Excel. One could argue that there were multiple ways to create reports and dashboards from the Microsoft database, including PowerPivot, but in terms of having a stand-alone product that makes for quick report creation that does not require fairly deep knowledge of SSRS and Visual Studio. Microsoft was lacking in the space, but that was filled within the last two years by the Power BI product. So, if you are coming at this topic as someone with Microsoft data platform experience, that does not include Power BI, it’s really difficult to have a point of reference.
What is Tableau
Tableau comprises of Tableau Desktop, a desktop report (let’s include terms dashboard and visualization) authoring tool that creates full-fledged reports that can run from within Tableau Desktop, and Tableau Server, a web based platform to host Tableau reports (and edit them in the web interface). In our current client’s project, which includes a self-service BI component, reports are typically authored in Tableau Desktop by report writers and self-service power users, and published to Tableau Server by someone from within centralized IT that can determine where to best publish and if the report meets compliance requirements (see: Project Governance of Self Service BI Without Killing the Self Service).
Tableau sits at the top, or last layer, of a BI platform – the visualizations presented to the user. As a product, it does not have the complexity involved with the installation and integration of the entire Microsoft data platform. Of course, a BI initiative that uses Tableau still requires a reporting database to use as a data source. But, Tableau is more loosely coupled and does not require someone with the broad knowledge required to install and make live the entire Microsoft data platform. One could argue that Power BI, in and of itself, can be installed independent of the rest of the Microsoft data stack. But it’s not my real-world experience to see it outside of a client, without a significant Microsoft data platform component.
Tableau Server is most like the cloud deployment platform for Power BI. However, specifically the product is on premise. When exploring how my client would have a “live” Tableau Server, I was able to find that someone had already created one in another area of the company. This was not an area where you would imagine the sophistication associated with Microsoft data platform administration. Installation, I was told, was accomplished in a matter of a few hours. Adding our area of the company, as users, was as simple as selecting a Microsoft Active Directory group to include in the security. When users connect to the Tableau Server URL from within a browser, Tableau Server detects who they are based on the LAN ID. Licensing was determined by user with an annual subscription.
Authors needed the authoring tool, Tableau Desktop, which compares most to Power BI Desktop. We determined who the authors would be, invited them to training, and included a link to install and a license key once they attended training. These authors included both “career” report writers and self-service BI authors from outside IT. We found that training the self-service authors went better than the “skeptics” anticipated. There is a true “drag and drop” simplicity to report development that seems more natural to me than with the Power BI alternative. Licensing required we purchase an initial installation license per user, then have an annual subscription per user.
For the Microsoft experienced, Tableau is most like Power BI. There are many differences in the way the two products perform, access data, the quality of visualizations, render geographic data, the simplicity of development, function syntax, installation, administration, just to mention a few items. Each product has their advantages, but that is beyond the scope of this blog post.
Want to learn more about Tableau or other business intelligence software? Contact us today: firstname.lastname@example.org