ITIL

What it means

ITIL is a set of best practices for managing IT services. It describes how an IT department or external provider should receive and resolve requests, manage incidents and problems, approve changes and measure service quality, so that services are predictable and meet the needs of the business.

It originated in the UK civil service and is now the most widely used framework for IT service management. The current version, ITIL 4, focuses on value for the customer and connects classic processes with agile approaches and DevOps instead of insisting on strict adherence to prescribed steps.

ITIL is neither a standard nor a piece of software. It is a framework from which an organisation picks what helps it. Individuals can be certified, not companies; organisational certification is done against ISO/IEC 20000, which draws on ITIL ideas.

How it works in practice

The element most often adopted is a single helpdesk or service desk, where every request is logged as a ticket with a priority and status. A distinction is made between an incident, an outage to be restored as quickly as possible, and a problem, the hidden cause of recurring incidents.

With an external IT provider, ITIL finds its way into the contract through an SLA: response time, resolution time by priority, service availability and regular reporting. Without such definitions it is hard to judge whether IT outsourcing really works and what the company is paying for.

Change management ensures that changes to systems are assessed, tested and scheduled in advance so they do not cause outages. In a small company a simple rule may be enough: no changes to the accounting or payroll system during the month-end close, and always a way to roll back.

What to watch out for

The biggest risk is bureaucracy. If ITIL is applied literally, even a minor request passes through several approvals and people start bypassing the system. The framework needs to be scaled to the company and checked for whether its processes actually speed up problem resolution.

Measure outcomes, not ticket counts. Metrics such as the number of closed requests encourage closing tickets quickly without solving anything. It is better to track user satisfaction, repeat incidents and compliance with the SLA, ideally as part of the provider's KPIs.