New Relic One provides an Alert Condition Recommendation service that uses AI and machine learning to recommend specific metrics and signals to monitor for entities in a production application. This service helps users figure out what to monitor and set up thresholds that are sensitive enough to catch anomalies but not so sensitive that they trigger distracting false alarms. The recommendations can be easily added to APM entities without alert coverage, and the model uses tags associated with an entity to generate recommendations based on the quality of the tags. The service also provides dynamic baseline alerts with recommended threshold values, allowing users to choose between default or custom thresholds. By increasing alert coverage, New Relic One helps users maintain a high level of system health visibility.