Model design best practices - Power BI Tutorial

- [Instructor] A good data model has the following benefits. The first is that you're able to easily navigate through the data. You're able to quickly find the tables that you're looking for, the columns that you're looking for. You're able to have a quick look, for example, at the model view and identify the relationships that may need to be edited. So think of the ability to navigate through regardless of the number of tables very quickly and easily. Next, your calculations such as your aggregations or the different types of decks, calculations that you'll be performing. The formulas themselves are simple, easy to understand and the running of these calculations are going to be as efficient as possible. And this has a big part to do with the types of functions being used. There are often substitutes for the functions that perform better than others. And this is going to take, of course, research. You're going to use the tools like Performance Analyzer to understand how quickly the visuals are loading and then you'll be able to make the necessary changes. Third, you'll have continuity when working in a team or on a project. It's always important to think about how things will extend past just yourself. So this is definitely a best practice. Fourth, reports are easy to build and are more accurate. Of course, the more accurate the information, then the right decisions can be made. And fifth and final here, troubleshooting is less painful. Now, following model best practices doesn't guarantee that you'll have no issues, nothing that you'll need to troubleshoot, but it does make the process now simple and less painful or simpler rather, and less painful when you get to that point where you have to troubleshoot if you are adhering to best practices, right? So these are just some of the things that you want to think about. And overall, a good data model is going to be relative. And so there are, you know, guidelines or best practices to think about and that way you'll ensure that you are on the right track. Now in terms of, you know, ways to think about it based on the scenarios that may be asked on the exam, anything that points to simplicity, efficiency and accuracy usually means you're on the right track. For data models, it's important to only import tables and data needed to tell the story. Think of it as writing a story with as few sentences as possible without losing the meaning of the story. So if the story can be told in 10 tables versus 30, then only import 10. Same for the columns within those tables. The concept of simplicity plays a role in how well your model performs and your user experience.