When you create a new column in a database table, you are redefining the schema that every query, join, and index relies on. It is fast if planned well, dangerous if done blindly. In production, a poorly executed schema change can lock tables, slow queries, or cause downtime. The impact is instant and global.
The method you choose depends on scale, database engine, and migration tooling. For SQL databases, ALTER TABLE ADD COLUMN is direct, but behavior varies across Postgres, MySQL, and others. In Postgres, adding a nullable column with no default is almost instant. Adding a column with a non-null default rewrites the table, which can be slow on large datasets. In MySQL, even a simple column add may trigger a full table copy without ALGORITHM=INPLACE.
For zero-downtime schema changes, consider phased migrations. First, add the new column as nullable, backfill it in small batches, then enforce constraints. Use migration frameworks or specialized tools like gh-ost or pt-online-schema-change for big tables. Monitor replication lag, lock wait times, and transaction duration throughout the process.