ITIL Change Enablement: Change Types, Risk Assessment and the Change Authority

Most outages are caused by changes: a new release, a configuration tweak, a patch. Change enablement maximises the number of successful service and product changes by ensuring that risks have been properly assessed, authorising changes to proceed and managing the change schedule. Notice the word enablement: ITIL 4 wants change practices that help good changes happen quickly, not just block risky ones.

What Is a Change?

A change is the addition, modification or removal of anything that could have a direct or indirect effect on services. That includes software, hardware, networks, documentation, processes, supplier arrangements and even contracts. It does not include the day-to-day work of running a service, such as restarting a failed process.

Change Types

TypeDescriptionApprovalExample
StandardLow risk, well understood, repeatable, with a documented procedurePre-authorised; no approval each timeInstall approved software on a laptop
NormalNeeds assessment and authorisation, scheduled in advanceChange authority depends on riskUpgrade a database version
EmergencyMust be implemented as soon as possible, usually to fix an incidentFast-track authorityPatch a critical security flaw being exploited

Standard changes are the biggest lever. Converting frequent, low-risk normal changes into standard changes frees the change authority to concentrate on risky ones.

Change Authority and the CAB

A change authority is a person or group who authorises a change. ITIL 4 recommends that authority be delegated to the lowest level that can judge the risk: a team lead for routine changes, a head of service for larger ones, an executive for the largest. A change advisory board (CAB) is one way to decide on high-risk changes, but it is optional, and holding a weekly meeting to approve everything is a common cause of slow delivery. Where teams deploy often, automated checks, peer review and pipelines can provide the control that a meeting used to.

The Lifecycle of a Normal Change

  1. Raise: record the request with a clear description and reason.
  2. Assess: evaluate impact, risk, dependencies, resources and test results.
  3. Authorise: the appropriate change authority approves, rejects or asks for more information.
  4. Schedule: add to the change calendar, avoiding freezes and clashes.
  5. Implement: carry out the change, with a rollback plan ready.
  6. Review: check that the change achieved its aims and note lessons.

Risk Assessment Questions

  • How many users and services could be affected if it fails?
  • Has it been tested in a representative environment?
  • Is there a tested rollback plan, and how long does it take?
  • Has a similar change succeeded or failed before?
  • Does it depend on another team, a supplier or an outage window?
  • Is the timing sensible (not month-end, not a holiday weekend)?

Worked Example

A team wants to upgrade the firewall firmware. They raise a normal change describing the version, expected five-minute outage and rollback to the previous image. The risk assessment records that the upgrade was tested on the standby firewall, that the primary will fail over to it, and that the date is outside the finance month-end. The head of infrastructure authorises it. The change runs on Saturday at 2 a.m.; monitoring shows traffic normal, and the team records a successful review on Monday. After the third identical upgrade, the team asks to turn it into a standard change with a fixed procedure.

Metrics

MeasureWhy it matters
Change success rateShare of changes that met their objectives without causing incidents
Incidents caused by changeShows whether assessment is effective
Emergency change percentageA high figure suggests poor planning
Lead time from request to implementationShows whether the practice enables or obstructs

Common Mistakes

  • Sending every change to one weekly CAB regardless of risk.
  • Overusing emergency changes to bypass assessment.
  • Approving changes without a rollback plan.
  • Failing to review completed changes, so lessons are lost.
  • Measuring approvals processed instead of successful outcomes.
Key takeaways
  • A change is the addition, modification or removal of anything that could affect services.
  • The three types are standard (pre-authorised), normal and emergency.
  • Delegate authority to the lowest level that can judge the risk; a CAB is optional.
  • Aim to enable fast, safe change, and track success rate and incidents caused by change.

Related Lessons