Expert Reference Series of White Papers Mastering Problem Management 1-800-COURSES www.globalknowledge.com
Mastering Problem Management Hank Marquis, PhD, FBCS, CITP Introduction IT Organization (ITO) leaders often confuse Incident and Problem Management, leading to more service disruptions of longer duration. Here we explore how an ITO leader can reduce the quantity and duration of service disruptions with Problem Management. Problem Management starts with Incident Management. Incident Management and Problem Management are not the same, although they are interdependent. If you do not have both, then you most likely have more user downtime, lower customer satisfaction, and higher costs than you should, all of which reduce ITO Return on Investment (ROI). Incident Management aims to restore service quickly by reducing the duration of disruptions. Problem Management seeks to prevent service disruptions by discovering the causes (or potential causes) of disruptions, and creating workarounds and permanent resolutions to them. These two can reduce both the quantity and duration of disruptions significantly. However, if you do not make the goals of each a priority, they tend to fall off in response to the day-to-day pressures of supporting users. It is just as important for you to allocate staff and require activities to develop workarounds, identify root-causes, and perform trend analysis as to restore service. If one functional group cannot do both well, then consider two groups but you must ensure both Incident and Problem Management activities get the management attention required to operate well. You simply cannot have one without the other, and they require tight integration with each other. Incident and Problem Management Depend on Each Other Incident Management is a one shot process that starts in response to a report of service disruption, and ends in service restoration. Its goal is restoration of service, and capturing information for use by Problem Management. Problem Management is an always on process, continuously examining information from any source that has or could initiate an Incident Management cycle. Its goal is prevention. Properly focusing on the objective of each (restoration vs. prevention) can result in higher service quality, increased customer satisfaction, and improved ITO ROI. Copyright 2010 Global Knowledge Training LLC. All rights reserved. 2
If you do not have both Incident and Problem Management, you can experience a higher number of longer-duration outages. Without Problem Management, ITO staff is doomed to fix the same issues repeatedly. Without Incident Management, ITO staff has limited data for analysis and cannot focus on Root Cause Analysis (RCA) and other prevention activities. One without the other usually results in more user downtime and steals valuable ITO resources away from efforts to add business value. To make tangible contributions to the success of your firm, you need to improve efficiency to free resources for other business-aligned projects (innovation). Following discrete Incident and Problem Management processes decreases call volume and reduces outage duration. These improvements can shift the balance between innovation and Keeping the Lights On (KTLO) enough to produce visible improvements in ITO business contribution, while freeing resources for focus on adding value beyond basic operations. Steps You Can Take To Get Started with Problem Management To improve (or start) your Problem Management, you need to take the following steps: 1. Reassess your understanding of Incident and Problem Management. The former seeks to restore unplanned service disruptions quickly while the latter aims to prevent them. These are two very different, albeit related, objectives. Consider how you currently operate with respect to restoring and preventing. Understand the objectives of Incident and Problem Management and what it means to your firm and ITO. 2. Assume that currently you combine restoration and prevention activities with little formal management over either. In most ITOs, prevention efforts receive much lower priority than restoration efforts even though prevention can reduce restoration efforts significantly. It is also common to fail to collect accurate information from each Incident Management cycle. Seek and compare the opinions of ITO managers, supervisors and staff as well as customers and users to get to the truth. 3. Investigate current activities to validate your assumption findings. If warranted, attempt to document the average duration and number of outages. Also include user downtime if possible. Do you have management objectives of each? What percentage of effort do you expend on each? Is your team capturing the information required to reduce their workload? Are they actively working to reduce outage duration and number of outages? 4. Assess the capabilities of your staff. At a minimum determine if there are 1) methods in place to record all Incident Management details, 2) any formal method for developing workarounds to speed resolutions, and 3) preventative activities such as trend analysis using Pareto Analysis (a.k.a. the 80/20 rule.) More mature ITOs include formal RCA techniques such as CFIA, Ishikawa, Kepner-Tregoe and others. Copyright 2010 Global Knowledge Training LLC. All rights reserved. 3
5. Start a service improvement program that formalizes Incident and Problem Management as distinct processes. Whether or not these are carried out by the same staff at different times. In most cases, separate staff is best due to the differing skills and job requirements of each. Ensure Incident Management tracks and gathers information Problem Management will use for trend analysis software support tools here are critical. Be sure that you allocate enough time for trend analysis, and be sure that there is a process for gathering industry information to identify potential disruptions, e.g., others upgrading to XYZ have had similar issues as a result. Using Pareto Analysis with simple control charts can quickly identify the top issue focus prevention efforts on that issue for fastest and most visible improvement. Organizing for Problem Management Many IT Operations (ITO) leaders mistakenly think that Problem Management requires reorganizing departments, leading to lower rather than higher customer satisfaction. In the following section I explain how to organize for successful Problem Management. Organizational structure plays a significant role in your success or failure with Problem Management. Correct organizational structure is critical to your success, but you probably should not reorganize to achieve it. Confusing process (task descriptions) with performance (task execution) and organizational structure can result in failure. Since Root Cause Analysis (RCA) often involves multiple technology departments or silos; Problem Management is a federated process by definition. In other words, Problem Management is not typically a collection of resolvers. Instead, Problem Management directs the performance of technical specialists, support groups, suppliers, contractors, etc. Under the governance of a Problem Management process, a Problem Manager coordinates the work of crossdepartment resources. Instead of a dedicated department, appointing a Problem Manager dynamically and federating resources often yields better results. A federated approach to RCA requires more coordination and information sharing than traditional RCA done by technology silo but it can also dramatically reduce the quantity, duration, and impact of outages. Done well, it also prevents outages from occurring in the first place. Process describes work, not organizational structures. While some organizational change to support Problem Management is inevitable, the real changes are not in where ITO staff sits or to whom they report, but rather how staff collaborates to solve a problem. Structuring for Problem Management process does not always require re-organizing departments, but it often requires empowering dynamic teams that know when, how, and why to perform. ITOs often confuse process and performance. Traditional ITO structures make little distinction between task descriptions (process) and task execution (performance). Research shows that often firms attempt to form two groups: Incident Management (triage) and Problem Management (root-cause analysis). The flaw in this approach can be increased outage duration, loss of organizational knowledge, damage to camaraderie, loss of management visibility, human resource issues, higher costs, and reduced communications. Copyright 2010 Global Knowledge Training LLC. All rights reserved. 4
A federated approach is often better than dedicated departments. One solution is a federated approach that leverages existing staff and capabilities as shared resources to form virtual organizations for resolving problems. This approach requires defined roles, responsibilities, and common oversight. From an ITIL perspective, this means creation and usage of diagnostic scripts, Incident and Problem Models, and Operational Level Agreements (OLAs). Steps You Can Take To Improve Problem Management Analyze your current approach to Problem Management (preventing service disruptions by discovering the causes of disruptions, and creating workarounds and permanent resolutions to them.) Understand what and how your Problem Management approach works today. There are four key indicators of Problem Management failure. 1. Incidents close too early and then recur. 2. Incidents remain open during RCA, losing visibility as to when service restores. 3. Missing agreed-upon Service Level Agreement (SLA) targets. 4. Having a large number of open incidents, many of which never close. Presuppose that currently you have one or more symptoms from the above list of failure indicators. Assume that Problem Management occurs as a series of disconnected activities primarily within a single technology domain or silo. It is common to have Problem Management activities within a single silo this is no usually as effective as Problem Management that spans multiple technology silos. If coordinated efforts spanning more than one silo seldom occur, then you probably do not have effective Problem Management. Seek opinions of others you trust in your firm to confirm. Investigate why Problem Management is silo-oriented versus federated. Does your IT Service Management (ITSM) support tool link and relate Incident and Problem Management records? Do second- and third-line resolver staffs have good working relationships with first-line Incident Management staff? Do all responders and resolvers regardless of silo understand business impact? Do you have Configuration Models, Problem Models, OLAs, escalations and related knowledge, and information readily available? Does all resolver staff receive regular training on the business implications of the services they support and the processes they follow? Start federating Problem Management now by creating a new role titled Problem Manager to be a single point of coordination and owner of the Problem Management process for the problems assigned. Work with your team to create escalation and integration with Incident Management. Create and agree-to OLAs with shared prioritization that empowers a Problem Manager to assemble resources as required. The Problem Manager coordinates activities; one or more technical support staff and/or vendors actually solve the problem. The Problem Manager uses OLAs and SLAs as well as escalation and prioritization as a model to engage additional resources when and as needed. Finally, the Problem Manager releases resources back to his/her organization, ensures knowledge management tasks complete, and then he/she returns to his/her normal function and role (until the next problem). Copyright 2010 Global Knowledge Training LLC. All rights reserved. 5
Summary Problem Management is much more than simply resolving an outage. Its reactive and proactive tasks span multiple departments and even incorporate suppliers and vendors. By understanding and separating Incident and Problem Management, you can reduce the quantity and duration of service disruptions. Realizing that, unlike Incident Management, Problem Management does not require a functional organization, is key to your success. Problem Management is pure process and designed for federation. Focusing first on Incident Management, and then Problem Management can lead to higher customer satisfaction through improved service quality, as it improves working conditions within your ITO. Learn More Learn more about how you can improve productivity, enhance efficiency, and sharpen your competitive edge. Check out the following Global Knowledge courses: How to Master Problem Management ITIL Service Lifecycle: Service Operation ITIL Service Capability: Operational Support and Analysis For more information or to register, visit www.globalknowledge.com or call 1-800-COURSES to speak with a sales representative. Our courses and enhanced, hands-on labs and exercises offer practical skills and tips that you can immediately put to use. Our expert instructors draw upon their experiences to help you understand key concepts and how to apply them to your specific work situation. Choose from our more than 1,200 courses, delivered through Classrooms, e-learning, and On-site sessions, to meet your IT and business training needs. About the Author Hank holds advanced certification in ITIL, ISO-20000, COBIT, Six Sigma and Project Management. He is a senior member of the American Society for Quality, and has over 25 years of IT experience managing, organizing and optimizing IT infrastructures and organizations. He has extensive experience helping IT executives implement IT governance and operational frameworks. He held operational management positions at MCI, was CIO at e-commerce financial services provider Celexis, served as VP of Marketing and CTO at management software company Opticom and rose to senior management at Compuware. Hank has helped dozens of companies develop and implement ITSM best practices. He recently led the Business Service Management practice area for Global Knowledge. Copyright 2010 Global Knowledge Training LLC. All rights reserved. 6