For the last day of Dashboard Week DS17 finally got the task we have been waiting for – create a dashboard using a different software, particularly – Power BI. The data was about grocery store sales from a real grocery store, available from Kaggle. This time we had only 4 hours on data prep, dashboard design, and blogging. Enough time to learn a new software from scratch, right?
The data came well structured and quite clean, so there was not much prep required. I only split the Date field into Date and Time fields and added new calculated fields for Profit and Total Sales per transaction. For this, I used Power BI Queries Editor that can be accessed via the main menu.
It was not clear from the original dataset what exactly fields PRICE, PRICEBUY, and PRICESELL mean and why there were so many zero values in the rows, but as there was no way for us to clarify it within the time we had, I had to accept the quality of the data set as it was.
Visualisation in Power BI
I had some experience of working with Power BI before, but it still took me a while to figure out how to create a chart and even more so, how to format it. My dashboard is a simple overview of sales and profit over time and per category.
I found Power BI’s interface less intuitive compared to Tableau, especially because many options are hidden away behind tabs or collapsed menus. There are many things in the software that take much longer to do than in Tableau, especially formatting of charts, and some functionality is not available at all (or require some programming skills). At the same time, I really liked the automated highlights and filtering functionality in Power BI, and the Switch Theme option that formats colours and fonts on a dashboard in one click. I wish there was something similar in Tableau.
Link to my interactive dashboard on the Power BI Data Stories Gallery.
Let me know if you have any questions.