I always felt that the report usage extracted from Power BI is just not enough. Data from “usage report” is nice, but is aggregated to daily level, while Activity logs need to be retained, because they will disappear. It is also not possible to extract e.g. slicer usage. We can do better!
For the last couple of months, I’ve been gathering real estate offers in a few polish cities. I’ve created a small dashboard to track prices of offers and also analyze trends.
This is a case that I have come across with one of my customers very recentely. I wanted to calculate sum of sales generated by first N products pareto style i.e. if first product’s sales is 200 and second’s 100, my calculation would show 200 for the first product and 300 for second. This allowed me to see that e.g. 90% of our total sales are generated by 7 products.
Today, I will go through how to create a loop function using List.Generate() and a Custom Function. I will show how to loop through pages in an API call. This would normally be solved using a for loop in e.g. Python.
I know that this case might be obvious for the most of the experienced data engineers and other specialists, but it really surprised me. Our Principal Architect in Clouds on Mars – Paweł Potasiński pointed me this possibility.
The case from one of my previous comments had another catch! As I mentioned the client was downloading data from an API. The client was interested in data for multiple entities i.e. had to use different credentials for the same URL to retrieve different datasets. Because the client wanted to see everything in one Power BI I had to find a suitable solution. Same as before, I turned my attention towards M language!
Today’s post shows how to call a custom function for each row in a table. This can be done both using standard Power BI interface and M language scripts. I have decided to try the latter.
Today I will show you a simple solution enabling the change of units/currencies or other perspectives using slicers. This will make a report more responsive and cleaner. From my experience, customers will be able to find the insights they are after much quicker. The end effect will look like this:
