Denormalization: when might you do it and what’s the trade‑off?

Answer

Denormalization duplicates data to speed up reads and reduce joins. It can improve performance for read‑heavy workloads, but increases storage, risks inconsistency, and makes writes/migrations more complex.