Here is a general idea how this feature could function: Add "degraded" outage level alongside "down". Allow the user to specify a customized amount of time that a website needs to be unreachable for in order for it to be classified "degraded" or "down", rather than reaching a final judgement of "down" after all the retries have been exhausted. This will most likely be implemented in a new major version release as it might require changes in the API revision/specification!