ITIL with Agile, DevOps and SRE: How They Fit Together in Practice
A common myth says ITIL is slow and bureaucratic while DevOps is fast and modern, so organisations must choose. In practice, most successful IT organisations use all of them. ITIL 4 was deliberately written to work with Agile, DevOps and Lean: its guiding principles, such as progress iteratively, collaborate and optimize and automate, echo those ideas. This lesson shows how the pieces fit.
What Each Approach Contributes
| Approach | Core idea | Strength |
|---|---|---|
| ITIL | Manage services to deliver value, with shared vocabulary and practices | Whole-service view: support, change, suppliers, service levels |
| Agile | Deliver in small increments with feedback and adaptation | Responsive development with close customer involvement (see What Is Scrum? A Simple Introduction for Beginners) |
| DevOps | Break down walls between development and operations; automate delivery | Fast, reliable, repeatable flow from code to production |
| SRE | Apply software engineering to operations, with explicit reliability targets | Measurable reliability and a way to balance it against new features |
Where They Overlap
- Iteration and feedback: ITIL’s “progress iteratively with feedback” matches Agile sprints and DevOps loops.
- Collaboration and visibility: ITIL’s “collaborate and promote visibility” matches DevOps culture and shared dashboards.
- Automation: ITIL’s “optimize and automate” matches CI/CD and infrastructure as code.
- Continual improvement: the ITIL model resembles retrospectives and the DevOps improvement cycle.
Practice-by-Practice Mapping
| ITIL practice | In a DevOps or SRE setting |
|---|---|
| Change enablement | Pipeline checks, automated tests, peer review and small changes treated as standard changes (see Change Enablement) |
| Release and deployment management | CI/CD, canary releases, feature flags (see Release and Deployment Management) |
| Incident management | On-call rotation, runbooks, swarming and incident commanders |
| Problem management | Blameless post-incident reviews with tracked actions |
| Monitoring and event management | Observability, dashboards and alerting tied to service objectives |
| Service level management | SLIs, SLOs and error budgets |
| Knowledge management | Runbooks and documentation kept alongside code |
SRE Terms in Plain Language
- SLI (service level indicator): a measure of how the service is behaving, such as the share of requests answered successfully.
- SLO (service level objective): a target for an SLI, such as 99.9% of requests succeed over 30 days.
- Error budget: the amount of unreliability the SLO allows. A 99.9% SLO leaves 0.1% as budget; about 43 minutes of downtime in a 30-day month.
Error budgets give teams a shared rule: while budget remains, ship features quickly; when it is used up, slow down and fix reliability. This turns the old argument between “move fast” and “keep stable” into a data-driven conversation. SLOs are close cousins of the service level targets in Service Level Management and SLAs, with a stronger engineering flavour.
Making Change Enablement Fit Fast Delivery
A team deploying twenty times a day cannot wait for a weekly board. The ITIL-compatible answer is to move control into the pipeline:
- Automated tests, security scans and quality gates act as the risk assessment.
- Peer review of every change gives a second pair of eyes.
- Deployments are small, reversible and monitored, so the risk of each is low.
- The pipeline records who changed what, when and with what test results, giving an audit trail.
- The change authority approves the pipeline and its rules, not each individual deployment.
- High-risk changes, such as major database migrations, still get human review.
Cultural Differences to Manage
| Tension | Way through |
|---|---|
| “Process slows us down” | Show where process removes rework; delete steps that add nothing |
| “Developers throw code over the wall” | Joint ownership: developers join on-call and post-incident reviews |
| “Operations always say no” | Agree reliability targets together and use error budgets |
| Different vocabulary | Build a shared glossary; accept that “incident” and “ticket” mean similar things |
Worked Example
A product team runs two-week sprints and deploys daily. The service desk team uses ITIL practices. They agree a combined approach: pipeline checks serve as the change assessment; feature flags separate deployment from release; the service desk receives release notes before each flag is switched on; support issues that repeat go to a shared problem board; and a 99.9% SLO with an error budget replaces the old monthly availability report. After three months, deployment frequency is up, incidents caused by change are down and the desk is no longer surprised by new features.
Common Mistakes
- Treating ITIL and DevOps as rival camps rather than complementary toolkits.
- Automating deployment but keeping a manual approval step that never rejects anything.
- Adopting SLOs without agreeing what happens when the budget is spent.
- Dropping service desk, knowledge and supplier practices because “we do DevOps”.
- ITIL 4 was designed to work with Agile, DevOps and Lean; their principles overlap strongly.
- Move change control into pipelines: automation, peer review and small reversible changes provide the assurance.
- SRE adds SLIs, SLOs and error budgets, which complement service level management.
- Choose the best of each approach and keep the whole-service view that ITIL provides.