Power View: Creating Maps

Power View, in addition to GeoFlow Power Maps, has the ability to take geospatial data and graph it to view data from a different perspective. You can’t create tours and some of the features that you can in Power Maps, but you can still demonstrate performance across a geographical location relatively simply. In this blog post, I’m going to take some geographical data around the Adventure Works database and graph it in Power View.

 

Initially, I’m going to create a connection to my Adventure Works tabular database and select Power View report as my import connection type. Under Geography, I’m going to select the StateProvinceName field, along with the Sales Amount field. My initial table looks like:

 

image

 

Up in the ribbon, on the Design tab, the fourth icon from the left for my report is the Map icon. When I click this, Power View is going to open up a connection with Bing maps, and attempt to graph my data geographically. Clicking on it, I get a very simple map showing my sales amount by State or Province around the world.

 

image

 

There are now a couple of different options that I can do. If I look at my field list, I can tile by something, create vertical or horizontal multiples, or drop in longitude and latitude, if I have it for example. I can also create a pie chart in each State/Province showing the breakdown of an additional category. If, for example, I take the Product Subcategory column and drop it in the Color section of the Field list, I get a map showing a Pie chart breakdown of each state/province’s sales:

image

 

Zooming in, I can see each countries breakdown by state or province. The only issue I see with this report is that everything is so spread out, it’s hard to see each country or region’s detail and keep the whole picture in focus. However, if I drag Region into the Vertical Multiples section, each Region becomes it’s own sub-map inside of the main Power View map:

 

image

 

There we go, much better. Now I have a nice visual Power View map that breaks it down but still keeps it easy to see the whole picture as well. One thing to note though, if I zoom in on France, there is some strange stuff going on with the province’s in the country. The same issue could be encountered with cities. The most accurate representation that we can use in mapping is Long/Lat, which if we had the fields available, should be what we use. If you don’t have Lat/Long (which, unfortunately, is more common than actually having it), we can pass concatenated field to Power View by setting them up in our model appropriately. The reason for this is that the same city, state, or province could happen multiple times throughout the world. How does bing maps know that we mean London, UK as opposed to London, Kentucky or London, Ohio? Back in the tabular model, I’ve created a concatenated field for “City State Country” and one for “State Country” by using the following DAX expressions:

City:

=[City]&","&[State Province Name]&","&[Country Region Name]

State:

=[State Province Name]&","&[Country Region Name]

 

Build and deploy, and refreshing my Power View model and replacing the old State field with my new State Country field fixes this inaccuracy issue and gives me a much clearer picture of, in this case, my sales in France, thereby leading to a much cleaner report:

image

 

In summary, while not as geographically amazing as Power Maps, creating maps in Power View is an excellent way to get some quick geographical reporting and visualization for a presentation, especially if the data model has already been built and is available in a tabular database or Power Pivot. However, when giving data to Power View to give to bing, ensure that you give Power View the most accurate geographical value of the data as possible.

Advertisements

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s