Skip to content
>devspeak
← glossary
level 4Databases

Normalization

also: normalize, normalized, normalise, denormalized, denormalization

Organizing a database so each fact is stored in one place, instead of being copied into many spots that can drift out of sync.

a bit more technical

Structuring tables to reduce duplicated data, typically by moving repeated information into its own table and linking to it with foreign keys. Denormalization deliberately duplicates some data for simpler or faster reads, at the cost of keeping copies consistent.

picture it

Keeping one contact card for your dentist, rather than typing her number into every appointment. When the number changes, you update it once and everything stays correct.

why it matters

Agents moving fast sometimes copy data wherever it's convenient. Asking about normalization helps prevent the classic bug where a change shows up in one place but not another.

what your agent might say
agent
The team name is duplicated in every member row. I'd normalize this into a teams table and reference it by team_id.
how you might use it
you → your agent
Check whether we store the same information in more than one table. If we do, explain whether normalizing it is worth a migration right now.
common mistake

Thinking any duplication is wrong. Some copying is deliberate, like saving the price paid on each order so later price changes don't rewrite history.

builds on