Here’s the draft section on grains1. It’s pretty short, and it might go into the entities or attributes section (or chapter). As always, feel free to poke holes in this one.
Thanks,
Joe Reis
When you model data, you’re inevitably faced with deciding the level of detail and granularity you’re trying to capture and represent. This level of detail and granularity is the grain of a dataset. And no, it’s not the grain of the edible variety. As you’ll learn, grain is one of the most critical choices when defining your data model.
You get data - and grains - in two ways: by creating new data or working with existing data. If you’re making new data, then a significant component of the data modeling exercise is determining the grain of the dataset. Otherwise, you’re dealing with the pre-existing grains of one or more datasets.
Keep reading with a 7-day free trial
Subscribe to Practical Data Modeling to keep reading this post and get 7 days of free access to the full post archives.