RDBMS design often emphasizes data normalization (especially 3 NF) for the sake of efficient transaction processing (OLTP). OLTP is the predominant use case for an RDBMS.
Goals of Normalization in RDBMS:
Free the database from modification anomalies
Minimize re-design when extending
Avoid bias toward any particular access pattern of an application
On the contrary, the single most important factor in designing the application database schema with MongoDB is about, matching the data access patterns of your application.
OLAP - Online Analytical Processing, aka reporting is generally performed on databases that have been de-normalized to facilitate OLAP as opposed to OLTP.
MongoDB not being an RDBMS,
Is MongoDB not recommended for OLTP processing?
Is MongoDB recommended for OLAP processing?