Official News from Microsoft’s Information Platform
Machine Learning Blog
A couple of weeks ago, Quentin Clark published “What Drives Microsoft’s Data Platform Vision?” In that post, Quentin outlined our work to simplify business intelligence. Here, I’d like to connect that introduction to our focus on Power BI for Office 365.
In my role at Microsoft, I demo software A LOT. In fact, showing off our solutions is one of my favorite parts of the job. My team and I love to show the capabilities of the product we are building—the ease of use, the fast and fluid interface, the power of collaboration in data analysis, etc. We deeply believe in our mission to simplify everything associated with Big Data and Analytics, and we think we have solutions that do exactly that! It is all about simplicity: to make BI/Analytics transparent and to enable everyday users to discover insights from data without having to learn new or complex tools.
From “OH NO!” to “Oh Yeah!!”
A few weeks ago, just as I was about to start an important meeting with a customer, I realized that I did not have my “demo laptop” with me. I started to panic…clammy palms and a nagging unease with “What am I going to do?” running through my head. The demo laptop is a big deal, you see. It has our entire “stack” installed, everything from Microsoft Office to SharePoint to SQL Server to a local instance of Hadoop, etc. Usually, no demo laptop means no demo.
This was a significant meeting with an important audience that had been difficult to schedule. Luckily, the meeting room had a PC with a browser (Chrome, actually) and a connection to the internet, and it turns out that this was all I needed for a great demo. All the components which had been set up previously on my laptop are now available online, as a simple yet powerful cloud service.
On the PC, I pointed the browser to my Power BI site, which I signed up for at powerbi.com, and as simple as that, I had access to everything required to demo how to identify and share some really interesting insights from a dataset. I searched for interesting data using Power Query. I quickly created a mash up and some visualizations in Excel—my favorite (and broadly familiar) data tool—and then published it to my Power BI Site on SharePoint Online in Office 365. The act of publishing the Excel workbook saved my work on the SharePoint site, which is provisioned and managed for me, and automatically moved the associated analytical model into Windows Azure.
A Little Background
To understand how exciting this experience was for me, it might help to have a little background. In the past, I—or someone on my team—would spend a few hours setting up a demo machine in preparation for a customer meeting or presentation at a conference. Essentially, we needed to maintain a “mini IT” capability on the team. Now, however, I am free of that dependency. I can dive into a feature-rich product demo with nothing more than an internet connection and a browser. Just like our customers, I can focus on the data, compelling visualizations, and business insights. I can show off the capabilities of our software and data tools without worrying about the infrastructure.
Another cool aspect of this experience is that when I published the Excel workbook to Power BI on Office 365, the solution instantly became a cloud-scale, multi-user, refreshable BI solution—no longer an island of data, but a manageable, reliable organizational asset. The solution renders in all modern browsers; it is device and platform independent, not even requiring a specific version of Excel on the consuming user’s desktop. Anyone with access to the Excel workbook through SharePoint Online, on any type of device, can work with the data and its underlying analytical model.
The Analytical Model
The analytical model is a core part of the Excel workbook and is incredibly powerful. The model describes the natural business language, relationships, and calculations that structure the overall solution. The model enables interactive and visual exploration and reporting of the data by business users—through Power View—without requiring a technical understanding of data modeling concepts like schema, primary and secondary keys, hierarchies, etc. Creating this model is as simple as—and, in fact, is initiated by—creating a chart or a PivotTable in Excel.
The embedded analytical model enables the Q&A capabilities of Power BI, which provide the ability to answer questions submitted using natural language. The business terminology captured in the model becomes the basis for the natural language interpretation by Q&A. Of course, the basic analytical model can—and usually should be—further annotated (through Excel) to strengthen the natural language interpretation capabilities. All it takes for business users to get insights on their data is to type in questions, similar to the way they ask a colleague a question. Power BI translates the natural language question into a technical query and returns results automatically formatted and visualized in the most relevant fashion. The best way to understand this is to see it in action; check out this video (made by Patrick Baumgartner on my team):
What Does It All Mean?
Power BI for Office 365 is a powerful solution that enables every business professional, regardless of technical skills or experience with data analysis, to explore data and uncover important, meaningful insights and to share these with others quickly and easily. This empowers business users who understand and run the business to make data-driven decisions without the usual complexities and dependencies associated with IT and BI systems. No special tools, no complex software installations, no hardware infrastructure to configure and manage. Just the simple, everyday tools of the business world: Excel, a browser, and natural business language. This is what makes Power BI so powerful; this is what cloud computing enables. So, give it a shot: go to PowerBI.com to learn more and sign up. Free trials are available now!
Kamal Hathi Director of PM Data Platform Group
Comments in this blog are open and monitored for each post for a period of one week after the posting date. If you have a specific question about a blog post that is older than one week, please submit your question via our Twitter handle @SQLServer