Should I design my schema denormalized from the start?

For a transactional system, no. Start normalized, because write correctness is hardest to retrofit and joins are rarely your first bottleneck. Denormalize specific hot paths when measurements prove the join is the cost. The exception is stores without joins, like DynamoDB or document databases, where you must design around access patterns, and denormalization is the design method rather than an optimization.

How do denormalized copies get updated when the source changes?

Three common mechanisms. Same-transaction updates keep everything consistent but make writes slower and more coupled. Database triggers automate it inside the database at the cost of hidden logic. Asynchronous propagation through change data capture or an outbox scales best but leaves a short staleness window. Most teams pair the async approach with a periodic reconciliation job that recomputes copies and repairs drift.

Is a materialized view the same thing as denormalization?

It is denormalization with the database doing the bookkeeping. The view stores a precomputed query result, so reads skip the joins, and refresh rebuilds it from the source tables, so drift cannot become permanent. The limits are freshness, since data is only as new as the last refresh, and refresh cost on large views. When those fit your needs, it beats hand-maintained copy columns.

Databaseschapter 4 of 5 · 3 lessons

Normalization vs Denormalization

Normalize to reduce duplication; denormalize to avoid expensive joins at scale.

← All chapters View in cheatsheet

A seller changes their shop name. Your database updates one row, and every page showing that name is instantly correct, because every one of them reads it from that row.

That is normalisation: each fact stored exactly once. It also means the page showing an order has to join four tables to show a name, a product, a price and a city.

At 50 requests a second those joins cost nothing. At 5,000 they are most of what your database does, and somebody proposes copying the seller's name onto the order row.

Do that and reads get fast and the shop name now lives in two places, so the next rename has to find both. Neither choice is a virtue. They are opposite ends of one dial, and which way it should turn changes over the life of a product.

Lessons

3 in this chapter

Normal Forms in PracticeOne fact, one place. That sentence is 90 percent of normalization.3 min
Why Reads Push You to DenormalizeJoins are cheap until traffic, data size, or sharding makes them the bottleneck.3 min
Keeping Duplicates HonestThe moment a fact lives in two places, you own the job of keeping them in agreement.3 min