A live, automatically-built diagram of which services call which, with health overlaid, the topological view of an incident's blast radius.
A service map (sometimes service graph or topology) is a live diagram of every service in a system and the calls between them, with current health overlaid. When an incident lands, the service map answers two questions instantly: which other services are affected (the blast radius), and which upstream service is most likely to be the root cause (the highest-out-degree degraded node). Modern maps are auto-built from telemetry, not maintained by hand.
An on-call engineer trying to diagnose a multi-service incident without a service map is doing it from a list. With one, they're doing it from a picture. The picture is faster, especially when the incident crosses team boundaries and the responder doesn't own every service involved.
See the part of the platform that handles service map in production.