Asked by rhn-support-sdodson on slack:

It's likely no surprise that we have testing gaps given the breadth of configuration options available. If we look to backfill automated testing for commonly configured items would CCX be interested in helping us understand which non default configs are most commonly used? Perhaps a report on most commonly configured values in config.openshift.io ?

An example provided:

an example is that we've recently found that we have no test that exercises the Image Registry configured to use cloudFront. So if we had a count of clusters with spec.storage.s3.cloudFront defined in configs.imageregistry.operator.openshift.io and that count were significant (perhaps > 5% of the fleet) we would use that as a trigger to ensure JIRAs were created to go add automated testing for that. So generate a list of all keys and a count of all clusters with keys that match? Then the Image Registry team could look at the report for configs.imageregistry.operator.openshift.io and make sure that they have test coverage for the top $x percentage of configs.
I don't think we'd need a lot of slicing and dicing capability present in the networking dashboard, but some easy hop from that count to a list of clusterIDs and then a generic report that took clusterIDs and produced a heatmap of common attributes like we do for CCX Symptoms alerts may be useful for scenarios where it does become interesting to know how often a given config is used on a given platform, cluster size, etc.

The goal for this task is to build a prototype report for the config values used in external clusters and iterate on it with Scott and other OCP eng representatives to get something useful out of the report and define further steps.