To answer this question, we need to talk about how where your data comes from (the "data set") and how Data Studio connects to it.
- Data set: the underlying platform that holds your data
- Data source: the schema (structure) of that data available to use in Data Studio
From Data Studio's perspective, there are 2 types of data sets:
Fixed schema data
Systems that store and aggregate their data in complex structures have a defined schema. (A data cube is an example of such a structure.) Google's measurement and ads products (Google Analytics, Google Ads, DV360, etc) fit this description. There are several implication here:
- Fixed schema data is well-defined. Data Studio knows up front which fields to expect and what the type and aggregation of each field is.
- Metric aggregation happens in the underlying system. For example, the Sessions metric in Google Analytics is a Sum, and Data Studio can't change that.Hence, the default aggregation for these fields is Auto.
- Fixed schema data is provided by each platform's API. Data Studio can only access the dimensions fields provided by that API.
Flexible schema data
Data that comes from systems where the user defines the structure is known as flexible schema data. For example, if you create a Google Sheet or MySQL database, you define the columns to suit your needs. Rather than 3 or more dimensional data structures, these system use simple tabular formats to hold the data. Connecting to flexible schema data has 2 notable implications:
- Data Studio can access all the columns (fields) in the data set.
- Numeric fields are unaggregated (default aggregation None). You get to define the aggregation you want for those fields.
OK, so why can't I use field X?
Certain fields in fixed schema data sets aren't actually fields at all: they are calculations performed by the front end product's business logic. For example, [need example]