This makes it easier to go through your list of flakey tests and fixing them.

It would be helpful if the flaky test information in Insights also included: A link to when the test first became flaky The ability to see flaky test information from a customized time period

Workflows that are rerun from failed should not be included in Insights, as they will often have a significantly different duration than a workflow that ran from start.

Adding a median duration column provides a more accurate indication of the duration users can expect when committing to a repository. It helps eliminate skews that creep into p95, but most importantly it helps folks working on CircleCI config see if their optimizations are driving impact.

I would like the ability to customize the date ranges for your insights tab. Essentially, the week, 30 day, 60, and 90 day options can be limiting.

I want to know which tests are failing the most across all branches, not just master. It's nice that Insights shows the most failed test in each branch but 99% of the time what I care about is which tests are flaky when I submit a PR. CCI-I-785

When orbs contain a job that is used in multiple repositories, it would be helpful if insights included aggregate metrics and test insights for the job cross-project. This would allow for easier detection of cross-project flaky tests and other potential issues.

When viewing insights for a particular selected branch or branches, insights for "All Branches" are still viewable. Allow users to remove these insights so that they can only view branch-specific insights.

In some cases, approval gates are used for rollbacks or circumstantial deployments. As an option, if all jobs prior to the approval are successful, then the workflow should be considered successful.