We used a metamorphic-testing-based approach to test implementations of hash functions, post-quantum candidate KEMs and signature algorithms, and found multiple interesting implementation bugs and some unexpected properties (of the schemes themselves). We used mostly known techniques and existing tools, but applied them based on our expectation of how a cryptographic primitive should behave. In this talk we share our experience, and encourage greater use of these (very accessible) techniques by describing how they can improve cryptographic implementations with limited effort.