The IT Infrastructure Library (ITIL) originated in the United Kingdom in the 1980s, with the goal of creating a set of standards for delivering efficient and high quality IT services. Twenty years later, growing complexity in business-critical applications, increasing demands from the business on IT organizations, and the pressure for regulatory compliance are converging to finally drive broad adoption of ITIL on both sides of the pond.

Forrester research estimates that ITIL adoption among billion-dollar companies will be increasing to 40 percent in 2006, and reaching 80 percent by 2008.

As companies rush to deploy ITIL, early practitioners often feel overwhelmed by the extensive process framework, unsure as to where or how to get started. Even organizations that have had a couple years of ITIL experience under their belt struggle to gather metrics to demonstrate their "ITIL ROI." The key to addressing both challenges is to take a closer look at two tightly related ITIL components, Incident Management and Problem Management.

Implementing Incident and Problem Management

By ITIL definition, Incident Management is focused on restoring normal service levels as quickly as possible with minimal disruption to the business. It's a highly visible process, and if done well, results in reduced service interruptions, increased efficiency and improved user satisfaction. A Forrester survey of companies larger than $1 billion showed that incident management is the No. 1 ITIL priority with an average ranking of 8.74 out of 10.

Problem Management takes a longer term view and is tasked with reducing the effect of incidents and errors, as well as proactively preventing them. A well implemented Problem Management system reduces recurring incidents and creates permanent solutions versus just one-time fixes.

Due to the nature of these two processes, Incident Management and Problem Management are usually viewed as the logical starting points for diving into an ITIL implementation.

Once you've selected a starting point, the next set of success factors are no different from most IT projects. Well trained staff and good process design, though crucial, are not enough to achieve this goal. A comprehensive tool set to support the process roll-out is the all important third component.

System Management products are not enough

Most large enterprise IT organizations use system management products such as HP Openview, IBM Tivoli, CA Unicenter, BMC Patrol, Mercury Business Availability Center and Microsoft Operations Manager to monitor their environment 24x7. Although this allows for proactive detection of incidents and does contribute significantly to goals of both Incident and Problem Management, it's far from enough.

Monitoring products are not able to actively diagnose root cause for complex applications, can't prevent the propagation of a single alert to a full blown "alert flood", and does not aid front line operations in actually resolving the issues plaguing the IT environment.

Service desk solutions can't succeed alone

Service desk software is the other class of software widely deployed in the ITIL arsenal. Some common solutions include BMC Remedy, FrontRange Heat, HP Service Desk or HP Peregrine Service Center. These products track incidents, problems and change processes, but also falls short when it comes to providing resolution capabilities.

In addition, despite of the array of reporting capabilities provided by service desks, many large enterprises are still unable to gather real intelligence to aid in their incident and problem management efforts. This is because service desk packages still rely on human operators to input the correct data; ticket names, descriptions and categories are entered by IT staff who often fail to follow the entry standards or have time to properly update tickets after manual tasks are completed.

Leverage Incident Resolution to bridge gap

Somewhere between incident detection via System Monitoring and incident tracking with Help Desk tools, there lies a major gap: the process of triage, diagnosis and ultimately resolution of these incidents is still a repetitive, manual, error-prone process.

Most alerts get escalated all the way to tier three, and any expertise around resolution tends to be poorly documented tribal knowledge. IT professionals constantly operate in firefighting mode rather than proactively addressing problems and automating their response to incidents. In summary, the ITIL goals of minimizing service impact and preventing recurring incidents largely remain unmet.

The good news is, after years of relying on ad-hoc, homegrown methods, enterprise IT organizations finally have a viable alternative - software solutions that automate the critical step of incident resolution. Several early movers have released products to tackle this space, and have gained significant venture backing to bring to market sophisticated solutions that helps to round out the ITIL tool set.

This new class of software automates complex diagnosis and repair tasks against the entire environment. These range from simple tasks such as restarting a server, checking network connectivity and changing system configuration, to complicated, nested procedures such as running a full set of diagnostics tests against a clustered J2EE application environment, or interrogating backed up MQ queues and automatically routing the problems messages to the correct location.

Early adopters of Incident Resolution software include the likes of Alaska Airlines and NYK Logistics. Combined with existing tool sets, each has realized significant ROI in their ITIL implementation efforts.

Dean DuVall, Alaska Airlines managing director of customer services, said, "An automated, repeatable problem and incident management solution that scales with our business has allowed us to empower our first level resources and offload work from our second and third level support teams. These teams can now focus more of their energy on non-routine issues and strategic tasks to improve overall service levels."

Bring it full circle

Together with system monitoring and help desk tools, Incident Resolution products result in rapid, efficient resolution and complete the process loop. Monitoring solutions proactively detect any service outage and issues an alert.

The alert automatically triggers diagnosis and repair procedures in self-heal or visually guided modes. The incident resolution product finds the root cause, resolves the issue and records relevant data points at each step into the accompanying ticket to provide an audit trail, as well as closes the ticket.

In addition to significantly cutting down the resolution time and meeting the goals of Incident Management, the increased incident and problem visibility provided by an automated solution helps tier two and three support teams with successful Problem Management. With automated data capture and enhanced reporting features such as ROI tracking, ITIL project teams can finally present to management a complete set of before and after pictures for key performance indicators (KPIs) such as MTTR, Service Level and operational cost savings.

Automated Incident and Problem Management solutions drive maximized uptime, lowered support costs and increased operational efficiency. These benefits come from empowering tier one support to perform more diagnostic and repair tasks and freeing up tier three support for proactive management. They are also the result of reducing the initial ramp-up time for training new support engineers and capturing all tribal knowledge into a centralized knowledge database. In addition, the IT organization benefits from better overall process control and a well-documented audit trail for Sarbanes-Oxley.  

Looking to the future

Once you've gained success in your Incident and Problem Management projects, you can branch out to Change Configuration, as well the other ITIL Service Management and Service Delivery modules. The good news is that the automation solution you've adopted for Incident and Problem Management will also be applicable in the other areas, and contribute to your overall ITIL success.