It's also about DRY and accurately modelling the data for general operations. Imagine the difference between a data environment where a GDPR deletion request comes in, and everything has a relation back to customer identity, and one where the customer identity is denormalised out to many places or only implicitly present.
Comments
As others mentioned, normalization is generally about saving space not increasing performance.
That said I’m 100% positive that all those key value pairs are indexed for searching and querying purposes.
Not so much indexed, just columnar.
It's also about DRY and accurately modelling the data for general operations. Imagine the difference between a data environment where a GDPR deletion request comes in, and everything has a relation back to customer identity, and one where the customer identity is denormalised out to many places or only implicitly present.