As if Tableau 8 wasn’t enough – today we had to deal with PowerBI. I’d never used the software before so this was a trial and error exercise really.
The Data and data prep
The dataset for today was a dataset from Kaggle that recorded all the transactions from one grocery store between 2016 and 2019 and the task was to create a KPI dashboard for this data.
The dataset (unlike some days this week) was quick to download and pretty clean and easy to use. But as I’ve learnt this week – filter down on a topic and filter down fast! So I chose to focus on avocados.
I prepared the data in Alteryx, filtering on avocados, parsing the date and creating a couple of fields. As I didn’t know much about how to use PowerBI or if I could create calculations within it, I wanted to create the fields I needed to Alteryx first so that I only needed to concentrate on visualising in PowerBI. So, in Alteryx I added a field for profit per transaction as well as cost per transaction and total sales for the item in the transaction.
In Alteryx, I also deleted a couple of fields I knew I wouldn’t be needing. Confusingly there seemed to be a couple of fields which could be mapped to the selling price of the item – there was a column for both ‘Price’ and ‘Sell Price’ so I deleted one of these as well. Not sure if it was the right one…
![](https://www.thedataschool.co.uk/content/images/wordpress/2020/01/Annotation-2020-01-10-124418-1024x352.png)
Visualising in PowerBI
At first PowerBI felt strange as I’m used to using Tableau, the software looks very different and things are organised very differently. However, after a bit of exploring and experimenting, I found that I could put together a KPI dashboard in the software pretty quickly which was handy because I only had a couple of hours.
![](https://www.thedataschool.co.uk/content/images/wordpress/2020/01/Annotation-2020-01-10-135954-1024x570.png)