The code smells that make your Cypress tests flaky, and how we fixed ours

I've spent more hours than I'd like to admit staring at a green-then-red-then-green Cypress test, trying to figure out what went wrong. Not the interesting kind of debugging where you learn something. The soul-crushing kind where the test passed locally, failed in CI, passed again when you re-ran it, and now nobody trusts the pipeline. That's the real cost of flaky tests. It's not the wasted minutes. It's the moment a developer sees a red build, shrugs, and says "probably flaky" without checking. Two weeks later, the bug that test was supposed to catch is in production. At BetterQA , we maintain Cypress suites across dozens of client projects. Fifty-plus engineers, each bringing back field reports from different codebases, different frameworks, different flavors of chaos. After a while, you start seeing the same patterns. The same handful of code smells show up in almost every flaky test. Here are the three that account for most of the damage, and what we actually did about each one. S