Choosing between WebDriverEventListener and java.util.logging for low-noise test alerts in Selenium 4

Goal: Configure Selenium test execution to emit meaningful alerts for user‑actions (clicks, finds) while suppressing the default INFO‑level chatter that obscures useful output.

Constraint: You can either attach a WebDriverEventListener to intercept specific callbacks and log only selected events, or adjust the java.util.logging level for org.openqa.selenium to capture warnings/errors without the listener overhead. Each approach trades fine‑grained control against runtime cost and potential loss of diagnostic detail.

Which mechanism provides the right balance for a CI pipeline that runs hundreds of tests per day, and how do you verify that the chosen method does not introduce measurable latency or suppress needed diagnostics? What criteria should guide the decision when the test suite includes both stable UI flows and flaky, dynamically‑loaded components?