Normalization vs Denormalization: The Trade-off

Welcome to **Day 94**. Today we discuss **Database Design**.

Normalization (The Rule)

This is what they teach in school: keep your tables small and specific. No repeated data. This is great for **Data Integrity** (prevents bugs).

Sometimes, your report is too slow because it has to join 15 different tables. In this case, you might decide to "Repete" some data.

Example: Storing the `customer_name` directly in the `orders` table, even though it's already in the `customers` table.

**Speed**: No join required to see who placed the order.

**Simplicity**: Your queries become much shorter.

**Data Drift**: If a customer changes their name, you have to update it in TWO places. If you forget, your data is now inconsistent.

**Normalize** your production app (where people are signing up and buying things).

**Denormalize** your data warehouse (where you build dashboards and reports).

Look at a complex join in your database. Would copying one column from Table B to Table A make it 10x faster?

*Day 95: Phase 7 Project—Optimizing a 1,000,000 Row Query.*