BigQuery's data changes look like any SQL, with one rule of its own: UPDATE and DELETE must have a WHERE clause. To change every row on purpose, write WHERE TRUE.
INSERT INTO t (cols) VALUES (...) or INSERT INTO t SELECT ... adds rows.UPDATE t SET col = ... WHERE ... changes rows; UPDATE ... FROM other WHERE ... takes values from another table.DELETE FROM t WHERE ... removes rows. (FROM is optional in BigQuery.)
MERGE upserts: update the rows that match, insert the ones that don't, and optionally delete.
MERGE practice.customers t
USING practice.updates s ON t.customer_id = s.customer_id
WHEN MATCHED THEN UPDATE SET email = s.email
WHEN NOT MATCHED THEN INSERT (customer_id, email) VALUES (s.customer_id, s.email);
Each statement's result says how many rows it changed. In this lesson, a copy of the customers table is ready in practice.customers.