For those of you who have been reading my blog for a while, you know my interest in PowerPivot, Power Chart, and Power View go back several years. (If you don’t, you have a lot of old blog entries to catch up on.) Anyway, I’ve been spending some time recently looking for what the next big thing is for data analysis. Several others have indicated that Power BI might fill that requirement but I spent some time looking around first before agreeing and picking up Power BI to see what it could do. I’m still getting acquainted with it, but I decided that I like it enough to take you along on my journey. So first I need to help you catch up a little.
The first thing you will need to do is to get a copy of Power BI Desktop installed on your local computer. Note that I said local computer, not your server. In fact, I’ve been running it quite successfully on a Surface Pro 3 with Windows 10 and SQL Server 2014. If you do not have SQL Server on your desktop but have it on a server that you can access, that will work just fine. For today though, you will not even have to worry about SQL at all.
To download and install a copy of Power BI Desktop, go to https://powerbi.microsoft.com/desktop.
You can also go directly to the download page at: http://www.microsoft.com/en-us/download/details.aspx?id=45331. You will see the system requirements on this page and you may notice that Windows 10 and SQL Server 2014 are not listed. That was all the more reason to try it and I can tell you that it appears to work just fine.
However you get to the download page, start the download and then install the application on your desktop.
Download and install the desktop. After the desktop is installed, it should start automatically. If it does not, go to your Start menu an entry that says Power BI Desktop. On Windows 10, this appears in a separate section along the left side under Recently Added. It may also create an icon on your desktop and even possible get added to your taskbar across the bottom of the screen in Windows 8 through 10. In some way, start it and you should see the following dialog at appears on top of the desktop application which fills in the background:
Along the left side are options to get data. We will see and explore some of the many ways you can get data from different sources over the coming weeks. You can also return to recent sources that you used. Next is a menu item that says, ‘Open Other Reports’. Desktops with their various charts, tables, and other visualization are referred to as Reports and are stored with a .pbix extension. If you had previous reports that you worked on and saved, they would also appear in this section allowing you to click on them to directly open them.
There are also several videos and tutorials listed in this dialog to help you get started. I strongly recommend that you watch these videos and perhaps read some of the tutorials. In fact, there are many more videos and tutorials for Power BI Desktop. Another good page to help you get started is the Microsoft Power BI Support page at: https://support.powerbi.com/knowledgebase/articles/471664.
Microsoft provides several sample databases and in future weeks, I will probably use the ContosoDW databases that I previously used for many of the PowerPivot blogs. But for today, let’s just try something to amaze your co-workers.
Almost everyone knows the site www.imDB.com. Using the new Edge browser, I went to the site and discovered that there are some new pages available. Most people use it to look for information about movies and the actors in them. The site also has information on television shows. In fact, it has two new pages of interest. One of the new additions is a page that displays the top 250 movies and another page shows the top 250 TV shows. Just because everyone does movies, I went to the top 250 TV shows page just to find out which year had the most highly rated shows. The following image shows a little of that page.
But what I’m really interested in is the URL. I can select the page URL from the Address box at the top of my browser. It should look something like the following:
Next I go back to my Power BI Desktop and click the Get Data icon in the External Data group of the home ribbon. Like many buttons, this one has a top portion and a lower portion. The bottom portion will open a dropdown of common data sources as shown below:
However, I can also click on the top portion of the button to display the following dialog box. This box shows categories of external data sources on the left and the names of specific data source types on the right. Since I want to get data from a web page, I need to specify a Web source. I could check each category just to see what was available and eventually find Web in the Other grouping as shown in the following figure or I could have selected Web from the previous figure. Either way, takes me to the same place. (But it is interesting to see all the possible data sources, isn’t it?)
After selecting Web, I am prompted to enter the URL of the web page that holds my data. This is where I past the URL that I captured previously when I displayed the page of top 250 TV shows.
When I click OK, Power BI analyzes the page and displays objects that might be something I’m interested in.
Of course the names like Table 0 and Tab1 have little meaning. However by selecting any of the objects, Power BI Desktop displays a preview of the contents of the object on the right side. This is how I ‘discovered’ that Table 0 contains the data representing the top 250 TV shows as shown below.
When I click load, Power BI Desktop loads the data into its local data model. In many cases, that model may contain columns that we do not need. It may also contain columns of concatenated data such as in this case in which the show ranking, show name, and show year appear concatenated together. We will need to fix this. Ultimately, I want to get to a table that looks something like this:
However, that will be next time. Have a great Labor Day weekend and get Power BI Desktop installed on your machines because next time we are going to start manipulating the data and creating visualizations.