Deployment complexity

Splitting deployables multiplies release coordination, compatibility matrices, rollback paths, observability wiring, data migrations, and environment drift. Each additional service needs CI/CD, secrets, health checks, and deployment ordering when schema changes span systems.

The trade-off is independent team velocity versus operational surface area. Fewer deployables simplify debugging; more deployables isolate blast radius if boundaries are real.

On interviews: walk through a safe rollout with backward-compatible schema changes, feature flags, and rollback when a downstream consumer lags.

Common pitfalls: deploying services and databases in the wrong order; no contract tests between producers and consumers; treating staging as production-shaped only on happy paths.

Checklist:

• Map release dependencies and ordering constraints.
• Plan backward-compatible migrations and dual-read phases.
• Define rollback triggers and data repair steps.
• Keep environments aligned on config and observability.