It matters because it is what makes managed service margins possible at all. Work that scales — patching, alerting, routine remediation — is work that can be done once and applied everywhere, and the tooling is what converts headcount into coverage.
The common misunderstanding is treating RMM as the service rather than the instrument. An agent deployed everywhere and monitored by nobody produces alerts into a queue no one reads, which is worse than no monitoring because it creates a record showing the warning arrived and was ignored.