What is Dimensional Modeling and why should you care?
The classic Excel PivotTable did a great job of letting us quickly pivot and slice data for years. There was a ton of logic built into the tool in order to make it easy for end users, but that ease of use had a cost: it didn’t scale to multiple data source tables. And worse, it actually kept you from learning the terms to scale your own knowledge.
At the core of every good Power Pivot solution is the Data Model. But do you really know how to work with it properly? What shape your tables should have? When you should split tables up, when you should flatten them, and how to manipulate the data on the fly to do so?
Course Description
Built by business professionals, for business professionals, this course is intended to teach you the right way to build solid and scalable dimensional models. Whether you are already building data models regularly, or are new to the concept of dimensional modeling, this course will give you all the tools you need to build the best self service business intelligence models possible.
After a review of the core benefits of the Power Query Data Model and reporting technologies in your favourite tool, we’ll look at some steps that you should consider when architecting your solution. These tips should help you crystalize what your audience requires, as well as clarify what data you need in order to get there. You’ll also learn how to identify if your data is “normalized” for consumption by the Data Model, and experience an example of cleaning up an ugly pivoted data set.
You’ll learn key concepts and terminology around data warehousing and dimensional modeling including Facts, Dimensions, Relationships, Schemas, Keys, and more. You’ll learn why the PivotTable has been “too helpful”, and how it can impact your ability to extend your data models.
Armed with the background theory on modeling, it’s then time to get your hands dirty with hands-on examples of solving several “many to many” join problems. From composite keys to bridge tables, slowly changing dimensions to flattening snowflakes, you’ll learn which tools and techniques to use and when.
With the theory and recipes behind you, we’ll then jump into one of the most exciting parts of the journey: looking at complex real-world cases that seem built to defy you. Not only will you learn how to apply dimensional modeling rules to solve the challenges, but you’ll also learn why – just sometimes – you may need to violate some of those recommended practice rules as well.
This is an exciting course, as it pairs two of self service business intelligence’s most powerful features (Power Query and the Data Model) together in one place, using each for what they were designed to do. You’ll leave armed with not only the experience, but handy reference cards to apply these techniques to your own data, and to determine, “Is this a Power Query job, or a DAX job?”
Who this course is for?
This course is designed for Data Professionals who have some experience with designing self-service business intelligence models in Excel.
Ideally, you should have encountered one or more of the following problems in the real world:
- Triggered a “Relationship Between Tables may be needed” error on an Excel PivotTable
- Been told you cannot create a relationship between tables because each column contains multiple unique values
- Received an error upon refreshing your Data Model because a column contains multiple values
- Created a relationship backwards in your Data Model
- Discovered a value that doesn’t seem to filter properly when drilling in to a PivotTable