Weekend Work 022 – Dates and Dashboards

YouTube player

Progress into the world of Power BI requires that you get used to features that are only within the Service, the most challenging of these is Dashboards. They don’t make any sense when you first look at them. They especially seem really limited as they are just tiles instead of the infinitely resizable shapes you get with Power BI reports.

Dashboards within the Power BI Service are, in essence, Menus. They allow you to link to many reports (and other web content). Structuring these menus adds a whole new dimension to your Power BI Applications. Remember, these should, in essence, be consumed by your customers and form a single pane of glass view across multiple datasets and reports that can exist within a Workspace. This is where they transcend the main Power BI Report content as Power BI reports can only exist based on a single Data Models; having a dashboard the pulls core metrics, say from HR, Sales, Manufacturing and Supply Chain makes a lot of sense and the only way to combine all that into a single view is with a Dashboard.

YouTube player

DateTime is something that you will see whenever you use system-generated data. However, the reason for it being a problem in Power BI often isn’t immediately obvious. In the video today, the point was to show that the increase in column cardinality made it extremely inefficient to store data within a DateTime column. At the same time, fundamentally, it does not explain everything. The next issue you would find if you wanted to use DateTime is that it would not join the calendar table easily. It is possible to build a calendar table with the right number of rows to work with the DateTime column. However, the amount of duplication makes it impractical. Most analysis is also about a day, week, month or longer, so the time component is essentially redundant for most calculations.

Leave a Reply