ITIL Incident Management: Priority, Escalation and Major Incidents Explained
When something stops working, users want one thing: to get back to work. Incident management is the practice that minimises the negative impact of incidents by restoring normal service operation as quickly as possible. It is usually the first practice an organisation adopts, and it is the one users judge you by.
What Is an Incident?
An incident is an unplanned interruption to a service or a reduction in the quality of a service. A server crash is an incident; so is an application that runs but takes thirty seconds to load a page. A failure of a component that has not yet affected any service can also be recorded as an incident so that it is handled before users notice.
| Record type | What it is | Example |
|---|---|---|
| Incident | Something broke or degraded | VPN keeps disconnecting |
| Service request | A routine ask for something standard (see Service Request Management) | Request for a new mouse |
| Problem | The underlying cause of one or more incidents (see Problem Management) | A faulty network driver causing all VPN drops |
| Change | A planned modification (see Change Enablement) | Upgrade the VPN gateway firmware |
The Incident Lifecycle
- Detection and logging: an incident is reported by a user or detected by monitoring, and recorded with a unique reference.
- Categorisation and prioritisation: assign a category and a priority.
- Diagnosis and investigation: check the knowledge base, known errors and recent changes.
- Escalation: involve specialists or management when needed.
- Resolution and recovery: apply a fix or workaround and confirm service is restored.
- Closure: confirm with the user, record the solution and close.
Priority: Impact and Urgency
Priority decides the order of work and is usually derived from impact (how many people or how much of the business is affected) and urgency (how quickly the incident must be resolved). A common matrix looks like this:
| Impact \ Urgency | High | Medium | Low |
|---|---|---|---|
| High | P1 Critical | P2 High | P3 Medium |
| Medium | P2 High | P3 Medium | P4 Low |
| Low | P3 Medium | P4 Low | P4 Low |
Each priority has target response and resolution times in the service level agreement (see Service Level Management and SLAs). Keep the number of priorities small; four is usually enough. A senior executive’s broken laptop is rarely high impact just because of who owns it.
Support Levels and Swarming
Many organisations organise support in levels: level 1 (service desk) handles common issues, level 2 (technical specialists) takes more complex ones and level 3 (developers, vendors) handles deep technical faults. ITIL 4 also describes swarming: instead of passing a ticket between levels, a group of people with relevant skills work on it together until it is resolved. Swarming works well for complex incidents because it avoids queue time at every hand-off.
Major Incidents
A major incident is one with significant business impact that needs a dedicated, faster response, such as an outage of the payment system. Organisations define criteria in advance and have a separate procedure with:
- An incident manager who coordinates and decides.
- A technical lead who directs diagnosis and recovery.
- A communications lead who sends regular updates to stakeholders.
- A conference bridge or chat channel and a timeline log.
- A post-incident review after resolution.
Workarounds and Resolution
The goal of incident management is to restore service, not to find the root cause. A workaround, such as restarting a service, using a backup server or switching users to a secondary portal, is a valid resolution if it restores the service. The permanent fix is the job of problem management.
Worked Example
At 9:10 a.m., fifty finance users cannot open the invoicing application. The service desk logs the incident, sets impact to high (a department is blocked) and urgency to high (month-end close) and so assigns P1. A search of the knowledge base finds nothing, so the incident is escalated to the application team, who notice a certificate expired overnight. They renew it at 9:45 and the service desk contacts the users to confirm. The incident record is closed with the resolution noted, and a problem record is opened to stop certificates expiring unnoticed.
Common Mistakes
- Using too many priority levels so nobody trusts them.
- Closing tickets without confirming with the user.
- Starting root-cause analysis before service is restored.
- Logging poor descriptions (“does not work”), which wastes diagnosis time.
- Having no major incident procedure until the first big outage.
- An incident is an unplanned interruption or reduction in quality of a service.
- The goal is to restore normal service as quickly as possible, using a workaround if necessary.
- Priority comes from impact and urgency; keep the scheme simple.
- Define a major incident procedure in advance, with clear roles and a post-incident review.