When evaluating whether to refactor a monolith into microservices, how would you assess if the issue is due to CI/CD tooling and build optimization rather than architecture?

Start by profiling the current pipeline: measure build time, test execution, linting, artifact upload, and deployment duration. Break the pipeline into stages and log each step’s duration. If the majority of time is spent in build or test phases, the bottleneck is likely tooling. Look for duplicated dependencies, large monolithic artifacts, or slow test suites that can be parallelized. Compare the cost of adding CI/CD improvements (e.g., caching, incremental builds, test sharding) against the effort of a full architectural rewrite. If the monolith’s codebase is tightly coupled but the business logic is stable, a refactor may be unnecessary. Also assess team velocity: if deployment frequency is low due to pipeline stalls, optimizing CI/CD can yield immediate gains. Only consider microservices when you see clear domain boundaries, independent deployment needs, or scalability constraints that cannot be solved by tooling alone. Document findings, propose incremental CI/CD changes, and measure impact before committing to a rewrite.