Implement Problem Management with These Seven Steps

Problem management is a continuous process. It encompasses problem detection, documentation of the problem and its resolution, identification and testing of the solution, resolution, problem closure, and generation of statistical reports. Many times in IT organizations, IT managers skip or mishandle one or more of these steps, which results in increased demands on the staff.

Here's a breakdown of the steps for managing problems. If you follow them correctly, you'll reap great benefits for your organization, including increased staff productivity and end-user satisfaction.

Step 1: Define problem management process and practices
The first step in establishing an effective problem management discipline is to publish a plan on how to handle problems. This plan should cover the following:

• Procedures for handling problems: What is done after a problem is detected and reported, how problem data is captured and stored, and how the problem is managed to resolution

• Roles and responsibilities of the IT support staff: Who receives the problem, who records all information, who handles problem resolution, and what each entity is supposed to do

• Measurements for problem resolution: What will be tracked to monitor the efficiency of the problem management discipline

• Problems to be handled and how to classify them: Severity and priority assignment methodology, and escalation guidelines

• Bypass procedures: Actions that can be taken to immediately restore system availability in the event of specific events or problems

Step 2: Detect or recognize the problem
In this step, you activate the necessary tools to detect problems. Use all facilities for capturing problem reports, including the help desk. Gather data and record all pertinent information in a location accessible to all support staffers. Notify affected users to help minimize the impact of the problem.

Step 3: Bypass the problem
As soon as the problem is detected, take all possible steps to bypass it or minimize its impact on users. Ideally, you should identify bypass procedures in advance, ensuring that they'll have no side effects on other systems, applications, or users. Keep in mind that a bypass is not a resolution of the problem. All too often, IT treats a bypass as a permanent fix, only to have the system eventually fail because the bypass was not designed to run forever or because the bypass affected other systems.

In some cases, IT managers use bypass procedures so often that they become the norm for "solving" the problem, when they actually do little to prevent the problem from happening again. Examples include rebooting a server or network router without identifying the source of the failure or pressing [Ctrl]