Creating a report or dashboard in Power BI is quite easy. A few mouse gestures and a couple of clicks and voila, we have a report!
Creating a report that tells a story at a glance, however, is hard. The ultimate test of a good report/dashboard is if someone without any prior knowledge can quickly understand it. Our audience should be able to see what the page is about and what each visual is about -- at a glance!
We also want to consider whether we have more "sizzle" than "steak". Is our visualization attractive but without substance? Don't get me wrong, "pretty" is important. We just need to find a balance between "pretty" and "useful".
Power BI has about 30 built-in visualization types. Add to that all the custom visuals that we can get at the Office Store and we literally have hundreds of visuals to choose from. What's a data analyst to do?
So here's a one-page cheat-sheet that I use all the time. It gives us a helping hand when trying to figure out which visual is best for the type of story we are trying to tell. Here are a few examples:
When we want to display a measure and compare it by magnitude, a Clustered Bar or Column chart will do the trick. Alternatively, a custom Bullet chart is very effective. But we probably shouldn't use a Gauge or Bubble chart.
|Trends over time are best visualized using a Line chart or custom Sparkline. However, an Area chart is probably not a good choice.|
|When we want to display a single value, use a Card or KPI visual.|
This cheat-sheet is a great starting point for telling your data story. Browsing the Power BI Data Stories Gallery is also a great way to get some inspiration. I'll also experiment by switching between visualizations until I find one that best suits the story I'm trying to tell. The good thing about using Power BI is that it can support this kind of "speed of thought" analysis.
Download this guide and keep it handy when you are trying to put some "steak" into your reporting "sizzle".
|We are overwhelmed by information, not because there is too much, but because we don't know how to tame it. -- Stephen Few|