I wish the user interface would be customizable to allow users to create personal context-specific workspaces to hide irrelevant data, rather than trying to have a one-size-fits-all interface. This would go a long way, as would introducing a usage-based pricing model for data ingestion, per GB of metrics or logs alongside a device-based model similar to New Relic, which would be more attractive for cloud-native companies with dynamic infrastructure. While dynamic alerting is great, the overall alerting system can be complex to configure. If LogicMonitor looks into going beyond the topology-based correlation to include AI that can group related alerts from different parts of the stack into a single probable cause incident, that will significantly improve the system.
