Recent Power BI Updates I Love

Traffic Light Conditional Formatting

Recent Power BI Updates I Love

Example of Traffic Light Formatting

Power BI is genuinely one of my favourite things in the world, it’s ability to allow anyone to view, and understand data is amazing and although there are others out there (Tableau, Qlikview, Hummingbird etc) I’ve found that Power BI is the easiest to set up and the easiest for users to pick it. Also as a Microsoft centered organisation it fits in well with our strategy of using Microsoft products and it’s ability to export to Excel and embed in Sharepoint make it an easy decision.

Below are a couple of the recent updates which I find really interesting and think are great additions to the product that will assist in moving the college more into a data driven approach.

Traffic Light Conditional Formatting

One of the most interesting recent additions is the addiction of “icons” within the conditional formatting for a visualisation, This Reg, Amber, Green or as we call it Traffic Light, system allows you to immediately see the figures you need to look at.

Allowing you, as per all conditional formatting, to specify the rules that affect these icons, and even to chose different icons or even up load your own, including little animated gifs, is a a huge step forward from what was there before which only allowed you to format text and background colours.

The options are full and easy to use

Python and R Visualisations

Another couple of recent additions that look like they could definitely be hugely useful are the ability to add Python and R visualisations making use of the coding within those languages to produce much more rich information and even to work with looking at predictive analytics from the data from within Power BI itself without the need to transform this data in an external system and them import it into Power BI.

This means that we can use the scipy stats module to calculate the correlation between two values and the linear relationship between them.

The example below is from the edX course “Analyzing Data with Python” and shows the correlation between Horsepower and Price in cars is statistically significant (something i’m sure we would all know, but it’s great to be able to prove this.

Calculation Correlation and linear relationship

There are a number of examples of use for this that i have in mind for work and i’m hoping to implement some of this in the near future.

  • Prev Post
  • Next Post