A data analyst is combining two feature columns for a customer scoring model: 'annual_income' (values ranging roughly 20,000 to 300,000) and 'support_tickets' (values ranging 0 to 15). The modeling team wants each feature expressed relative to its own mean and spread so that features on very different scales contribute comparably and are centered around zero. Which transformation should the analyst apply?