A further tool under the Writer menu is labelled Import Feature Types.
As with Readers, to understand importing feature types, it’s important to recognize that a dataset is made up two parts. One part of a dataset is obviously the data itself, but another - separate - part is the dataset schema.
What the import tool does is read the schema of a dataset (not the data) and use it as the template for a set of feature types in a workspace.
One reason to import feature types is to write to extra tables in a database.
For example, a user has a workspace that writes to a spatial database. It only needs to write to a single table (roads), so there is a single Reader representing the database, and a single feature type representing the table.
However, at some future point the user decides the workspace also needs to write to a second existing table (‘rail’). The simplest solution is to use the command Writers>Import Feature Types...
The import tool will take the schema definition of the database table ‘rail’ and add it to the workspace.