Loading page
About this tool
Define SEV levels with impact, response targets and escalation paths, and export a ready-to-publish Markdown severity matrix.
This builder produces an incident severity matrix — the table that maps SEV-1 through SEV-5 to customer impact, first-response targets, status-update cadence and escalation paths — and exports it as clean Markdown. It follows the numbered-severity convention used in Google SRE incident management and the Atlassian and PagerDuty incident handbooks, where SEV-1 is the most severe. Engineering managers, SREs and support leads can adapt every field to their own on-call reality and paste the result straight into a runbook or wiki.
Open Incident Severity Matrix Builder on AltFTool — it loads instantly in your browser.
Add your input to the workspace.
Adjust the options until the result looks right.
Copy or download the output and put it to work.
Starts from SEV conventions used in SRE practice — critical, major, minor, low — so you edit rather than invent.
Impact, examples, response target, update cadence and escalation are all customisable per level.
One click copies a formatted table plus per-level definitions for your wiki, runbook or repository.
A SEV (severity) level is a ranked classification of an incident's current customer impact, where SEV-1 conventionally means a critical, product-wide outage or data loss and higher numbers mean progressively smaller impact. The level determines who gets paged, how fast the first response must be and how often status updates go out.
Severity measures the impact of the incident right now; priority ranks how urgently the follow-up work gets scheduled. A cosmetic bug on the pricing page can be low severity but high priority, which is why mature processes track the two separately.
Three to five is the practical range, and this builder supports two to five. Fewer than three loses useful distinction; more than five forces responders to debate categories during an outage instead of acting. Many organisations only page humans for the top two levels.
The first responder assigns an initial severity based on observed impact, and the incident commander adjusts it as scope becomes clearer. The working rule in most handbooks is to choose the higher severity when in doubt and downgrade later, because under-calling an incident delays escalation.