What is test data management?

Test data management (TDM) involves creating, maintaining, and managing the data required for testing — ensuring tests have appropriate, isolated, realistic data without causing issues. Challenges: sensitive production data (PII — GDPR), maintaining data across test runs, performance with large datasets, data freshness, and environment consistency. Strategies: (1) Data factories/builders: programmatically create test data on demand using builder patterns — no hardcoded IDs, fresh data per test. Libraries: FactoryBot (Ruby), Faker (multi-language for fake data), Bogus/.NET. (2) Database seeding: pre-populate with a known starting state per test run. (3) Transaction rollback: wrap each test in a transaction, roll back after — no cleanup needed. (4) Anonymization/synthetic data: generate fake but realistic data instead of using production PII. (5) Test data API: expose endpoints in test environments to create/reset state. (6) Data virtualization: mask and serve subsets of production data. (7) Contract-based data: use Pact contracts to define the minimum data structure, test with that. Good TDM is a critical but often neglected part of test infrastructure.