The table stopped working the moment you added the new column.
You stared at the schema. The query that ran in milliseconds now crawled. The indexes no longer hit. The data pipeline broke in three places. A schema change is never just a schema change.
Adding a new column can be simple or catastrophic depending on how you do it. In relational databases like PostgreSQL or MySQL, an ALTER TABLE ADD COLUMN command can lock the table. On large datasets, this lock might freeze writes and block reads. On production systems, those seconds turn into downtime.
Plan the new column. Decide the data type first. Pick defaults carefully. Null defaults avoid backfilling but can require extra null-check logic. Non-null with defaults can rewrite the table. Understand your database engine’s exact behavior for column addition.
For performance, consider adding the column in a way that avoids rewriting all existing rows. In PostgreSQL 11+, adding a column with a constant DEFAULT and NOT NULL can be instant. In MySQL 8, ALGORITHM=INSTANT can skip the full table copy if conditions are right. Always test on a staging dataset before touching production.