A new column in a database is small in size but large in consequence. Schema changes touch live systems. They can lock tables. They can block writes. They can cascade into downtime that no rollback can clean up. That is why adding a new column should be deliberate, fast, and predictable.
Before creating a new column, check for query patterns that hit the table at high frequency. Assess indexes. Adding indexed columns can double or triple migration time. Avoid default values on large tables when possible, because they rewrite the entire table. Use database tools that run online schema changes to reduce locking. All of this makes the difference between a smooth deploy and a pager at 3 a.m.
In PostgreSQL, ALTER TABLE ... ADD COLUMN is simple but not always safe under load. MySQL with InnoDB can add columns instantly in some versions, but behavior differs across releases. Always confirm the behavior in your specific engine. Use staging environments with a dataset that mirrors production size before running the migration in production.