Impact Analysis turns "this might be big" into a precise map. For any active or past incident, see the affected services, the affected customers, the regions, the downstream batch jobs, and the API consumers that hit error rates above baseline. The map updates live during incidents. After incidents, it freezes for the postmortem.
Five orthogonal dimensions captured for every incident. Service: which services are seeing elevated errors. Customer: which named customers and how many anonymous users. Region: which AWS/GCP regions and which sales geographies. Batch: which scheduled jobs depend on the affected service and may fail their next run. Partner: which API consumers (third-party integrations) are seeing elevated 5xx.
During an active incident, the map updates every 30 seconds. New downstream services that start erroring appear as new edges. Customer impact climbs as more sessions hit the affected path. Use the live map to make scope decisions: page more on-call? Move to a major incident? Tell support to write a status page entry?
When an incident closes, the impact map freezes with the final blast-radius snapshot. The postmortem builder cites this snapshot directly: "blast radius covered 4 services, ~14k users, 6 partners." Numbers in the postmortem match numbers in the system. No "I think it was around 10k" guesses.
When the map crosses customer-facing thresholds, the page produces a draft status-page entry pre-filled with the impacted regions, services, and a calibrated apology. Edit, post, done. The draft uses the live map so it is accurate at the moment you press post.
Subscribe to Nova AI Ops on YouTube for demos, tutorials, and feature deep-dives.
Impact Analysis is the answer to "how many customers are seeing this?" already on screen by the time the page hits.