Power BI and maps…. what are the options?
At GeoRhythm we are all about getting insights out of geodata. A cartographic representation of data, I do not need to tell you this, is extremely suitable to get across information with a location component.
Using dedicated GIS tools, however, can come with quite a steep learning curve for geo newbies. Enter Power BI, the most ubiquitous BI tool of the moment. Other BI tools sometimes offer similar possibilities, but let’s stick to good old Power BI for the sake of simplicity.
- Maps as a standard visualization
In essence, maps are indeed just another type of visualization. Power BI offers two main built in map types: a bubble chart and a choropleth map. Both use Bing Maps as a background. To get the data on the map, a basic geocoding method is used which seems to be string based. People with geocoding experience will know that this is quite prone to error (see image below) but with some manual corrections it does the trick.
To simply show data on a map, these options will often suffice. Nevertheless, this offers not much room for customization or let’s just say it, for making interesting maps.
For the sake of completeness, you can also make shape maps based on imported GeoJSON. These maps lack a cartographic background and can therefore merely be categorized as maps and are more map-like visualizations.
Geocoding can by quite the challenge, Power BI map on the left, ArcGIS for Power BI on the right
- Let’s step it up a notch with ArcGIS
For the more ambitious cartographic visualizers, Power BI natively supports ArcGIS Maps. Even without extra license costs, this offers some additional options. To name some interesting options: drivetime calculations, clustered maps, heatmaps, …
The most interesting feature, if you ask me, is the possibility of adding additional map layers from the ArcGIS online reference library (often behind the pay wall, see below).
To pay or not to pay (source: doc.arcgis.com)
As a background map you can choose between, among others, OpenStreetMap and TomTom.
One further caveat is that Power BI dashboards with an ArcGIS map can only be correctly embedded or published if the data shown is publicly available, something to consider when making dashboards for a larger audience.
Making more complex maps comes does come at a cost, the cost being a steeper learning curve. Some people may find the interface (and logic) too complex.
- Are there more options?
Sure, there are…but let’s keep that for the next GeoRhythm blogpost, stay tuned!
