Severity Heatmap is the GitHub-style calendar of your incident history. Each square is one day, color-coded by the highest-severity incident that fired. Patterns jump out: end-of-month deploy spikes, post-launch weeks, the Tuesday-after-long-weekend correlation. Once you see the pattern, you can plan around it.
Each cell is colored by the highest-severity incident that fired that day. Empty cells (no incidents) stay grey. Sev-1 is red, sev-2 is orange, sev-3 is yellow, sev-4 is light green. The color encoding is the same one used across the rest of the platform so cross-referencing with other dashboards is one less context switch.
The page surfaces patterns automatically. Day-of-week aggregation shows which weekdays carry the most incidents. Time-of-month aggregation flags end-of-month deploy spikes. Time-of-day overlay (when you toggle hourly mode) highlights the Tuesday-3pm-after-the-weekly-deploy cluster. Knowing the pattern lets you plan around it.
The drill-in shows every incident on that day with severity, owner team, duration, root cause (if postmortem published), and a link to the war-room replay. The day cell's color story unfolds into the actual incidents that drove it.
Three patterns we see customers use the heatmap for: (1) post-deploy retros, where a sea of orange the day after a release plans the rollback policy, (2) seasonal capacity planning, where Black Friday's heat maps to next year's reservations, (3) weekly engineering reviews, where a single screenshot anchors the conversation.
Subscribe to Nova AI Ops on YouTube for demos, tutorials, and feature deep-dives.
The heatmap turns a year of incident data into one screen. Most teams find a recurring driver they did not know they had.