Bar Graph (1 measure by 1 dimension)
The bar graph is most common of chart types and is useful for almost any type of analysis. So, let's look at Profit by Region.
|Profit by Region (Bar) (Excel)|
|Profit by Region (Bar) (Power View)|
|Profit by Region (Bar) (Tableau)|
Bar Graph (1 measure by 2 dimensions)
Due to their interpretability, let's stay with bar graphs. However, let's crank it up a notch and add a second dimension. Let's see what Profit by Region and Category gives us.
|Profit by Region and Category (Bar) (Excel)|
|Profit by Region and Category (Bar) (Power View)|
|Profit by Region and Category (Bar) (Tableau)|
Winner: Power View
Scatterplot (2 measures by 2 dimensions)
Now we're getting to the neat stuff. How well can these tools handle dense graphs? Let's try plotting Sales and Profit by Region and Customer (A hierarchical relationship!).
|Sales and Profit by Region and Customer (Scatterplot) (Excel)|
|Sales and Profit by Region and Customer (Scatterplot) (Power View)|
This wasn't a very difficult task in Power View. We also like the aesthetics of the graph. All-in-all, this isn't a bad chart. However, one of our biggest concerns with Power View is that it shows representative samples when the amount of points gets large. This makes outlier detection nearly impossible. If there is a way to turn this off, we haven't found it. This being said, Power View doesn't seem to be a good tool for this type of chart. Let's see what Tableau offers.
|Sales and Profit by Region and Customer (Scatterplot) (Tableau)|
Mapping (1 measure by 1 geographic dimension)
Mapping is a newer type of technology that is becoming more mainstream. Let's see how these tools handle it. We'll try Profit by State.
|Profit by State (Map) (Excel)|
|Profit by State (Map) (Power View)|
This wasn't a complex task in Power View. However, the points can only be displayed as pie charts, or circles in this case. Also, the points can only be colored by a dimension, not an automatically discretized measure. This chart looks nice, but it's difficult to discern too much information from this. Let's see what Tableau can do.
|Profit by State (Map) (Tableau)|
Now that we've gone through a few different important chart types, it's become apparent that Tableau is the better choice for basic charting. We realize that Tableau is not easily capable of creating clustered bar charts, which we find extremely useful. Perhaps they will introduce this feature in a later version. Some of you might be screaming "What about Power Map?!?!" We purposely left it out of this analysis, which is slightly unfair. We wanted this to be "BASIC" charting. Power Map, formerly GeoFlow is an extremely cool tool that allows the user to do a large amount of things related to mapping. In fact, a colleague of ours, Jason Thomas, recently did a really cool webinar and a series of blog posts related to GeoSpatial Analysis in Power Map. There will be a section later that deals solely with mapping where we will introduce Power Map. Thanks for reading. We hope you found this informative.
Associate Data Analytics Consultant