The table waits. Empty. A single schema defines it, but it is not enough. You need a new column.
A new column changes the shape of your data. It can capture new metrics, store user preferences, log system states, or track events your product never measured before. In SQL, adding a column is a direct act: you alter the table, define its type, and enforce constraints. It is an atomic operation that can alter production workflows in seconds.
The simplest path is ALTER TABLE. You name the column, define its data type, set defaults if needed. Time matters—on large datasets, a blocking alter can lock writes. That’s why experienced teams plan these changes during low-traffic windows or use online schema change tools.
For NoSQL, adding a new field is often schema-less at the storage level, but schema enforcement can happen in application code or middleware. Even here, define expectations and validation early. Migrations are not just for SQL—they are patterns that protect integrity.