Adding a new column is simple in theory, but costly when done wrong. Schema changes can lock tables, block writes, or trigger full table rewrites. On high-traffic systems, that risk is real. The safest path starts with understanding how your database engine handles these changes.
In PostgreSQL, ALTER TABLE ADD COLUMN is fast for nullable columns without defaults. Adding a column with a default value pre-13 will rewrite the entire table; in 13+, it avoids the rewrite. MySQL behaves differently: InnoDB may require table copies for some column alterations, and performance impact depends on storage formats and indexes.
Plan the column’s data type and nullability before adding it. Small types use less storage and keep indexes lean. Avoid adding new columns with complex defaults unless you control traffic or run the migration during low load windows. For large datasets, break changes into phases: add the column, backfill data in batches, then set constraints or defaults.