Sunday, December 4, 2016

Quick Insights In Power BI

Quick Insights is a unique feature of Power BI that uses Machine Learning Algorithms to search for patterns, trends, and correlations in data, and provides a list of insights in the form of visuals. These insights (visuals) can be saved by pinning them to a dashboard. To run Quick Insights on your dataset, click on the ellipsis next to your dataset and then select View Insights as shown below.

Power BI will run Machine Learning Algorithms on your dataset to look for correlations, outliers, trends, patterns, etc. and a notification pops up at the top right-hand corner indicating that the insights are ready. This process might take some time depending on the size of the dataset.

 Clicking on the View Insights button will show all the visualizations that were put together by the algorithms, as shown in the screenshot below.

If you like a particular insight, you can save the visual by pinning the visual to a (new or existing) dashboard. Another interesting thing to note here is that Power BI lets us run Quick Insights on each of these visuals individually. For example, if I click on the first visual produced by Quick Insights (the visual will open in focus mode), there is an option called Get Insights on the top right corner as shown in the screenshot below.

 If I click on Get Insights, Power BI will run the Machine Learning Algorithms again, but this time, it just focuses on the fields used in this particular visual. As shown in the screen shot below, the algorithms produced a quick summary of the existing visual and a list of new insights based on the fields used in the current visual.

I can click on any of these new visuals and get new insights from them (insights from insights) and continue this process. The option to run Quick Insights is also available for individual tiles in a dashboard just click on the “focus mode” option on the tile to open in focus mode, and you’ll see the option Get Insights on the top right corner. 

I hope this helps; please use comments if you have any questions.

1 comment:

  1. Thanks for writing this in-depth post. You covered every angle. The great thing is you can reference different parts.Enroll with Power BI online Training
    Power BI Online Training