alphaspirit - Fotolia

The missing ingredient for effective problem management

Problem management implementations often fail or have limited success because they lack managers who are trained to lead problem investigations using structured methods

Many problem management implementations fail or have limited success because they miss one key ingredient in their practice: Trained problem managers leading problem investigations using structured methods.

This has been shown to be the primary difference between effective and failing problem management functions. By following a few simple guidelines, your problem management function can be successful from day one or rescued from its current low levels of performance.

Typical example

A typical problem management process document usually includes the following sections:

Introduction: What problem management is and does. How it is different from incident management is sometimes included as well.

Roles and responsibilities: The different players involved, usually subject matter experts or resolver teams, management, incident management, problem co-ordinators and (occasionally) the problem manager role. It is surprising how frequently the problem manager role is not defined at all. Responsibilities for the resolver group usually include “investigate root cause” and “update and close problems”. The problem manager is often given responsibilities like “assign problems to resolver groups” and “track problem progress”.

The process: This normally covers raising and assigning a problem for investigation, finding cause and a solution and closing problems. There is usually no reference to how to go about solving the problem. Several process descriptions I have seen specifically state “assign the problem to a resolver group for investigation” as well as to implement a solution or workaround. Usually the process includes that resolver group closing the problem without any referral back for review or quality control. This means there is no way of knowing if the root cause found is correct, if the solution is adequate or even if it has been implemented.

Governance: If there is a section in the document on governance, it is usually limited to a statement that “the process owner is responsible for process adherence”. Rarely or never included are any mention of training, a description of a standard method for finding root cause, how solutions are tracked to ensure they are implemented, or key performance indicators (KPIs) and operational metrics to measure whether the process is successful or otherwise.

The result is that many (I hesitate to say most, but success appears to be rarer than expected) implementations do not achieve their expected results. I call this approach “passive” or “administrative” problem management. One of my colleagues calls it the “distributed” model. The impact on reducing incidents is usually minimal.

Figure 1: Monthly occurrence of major incidents

If your monthly major incident data looks like that shown in Figure 1 above, you may have one of these typical implementations.

The alternative: Active problem management

The missing ingredient in a typical implementation is skilled problem managers using a consistent, evidence-based, structured approach to solving problems. By structured, I mean either to adopt one of the major problem-solving frameworks such as Kepner and Fourie and follow it consistently, with all problem managers using it the same way and all the time.

An alternative is to agree your own set of steps (I set out one version in my book) and again have all problem managers follow it consistently. Deciding on a standard method that everyone will use – with no exceptions – is the critical success factor for effective problem management.

The benefits are:

      • Speed to root cause – a standard approach yields results more quickly – around 60% quicker, in fact (see Figure 2);
Figure 2: Average time to find root cause in two problem management implementations
      • Consistency – all your problem managers can be equally successful;
      • Certainty that real causes are found – because investigations are based on evidence and not guesswork and theories, you can show that the causes found are correct;
      • Collaboration – if you do problem management the same way every time, teams know what to expect, they can see the good results and they get used to working together without confusion.

Skillset for problem managers

In problem management, like incident management, analytical skills are more important than technical skills. Like a good detective, problem managers need to be methodical, evidence-driven and very structured in their approach, working carefully through the evidence to arrive at what, on the balance of probabilities, truly caused an incident or might in the future. The main skillset is always a good knowledge of problem-solving techniques.

Technical knowledge is useful to give the confidence to challenge subject matter experts, particularly if they are invoking their deep technical knowledge to suggest that their opinion should be accepted without question. Problem managers should always seek evidence to support assertions, ensure that alternatives are properly assessed and that actions proposed are sensible (and that they really have been done when people say they have been). Having said that, it is entirely possible to be an excellent problem manager with no technical skills at all.

Problem managers lead investigation sessions

Because it is the problem managers who are highly skilled in problem-solving techniques, they should lead problem management investigations in conjunction with the technical experts, then work with subject matter experts to determine solutions to problems and track implementation to ensure the problem is entirely fixed. The problem management function should be responsible for reporting root cause, progress on resolution and all the metrics and KPIs related to problem management. It is also responsible for maintaining a consistent and structured approach to problem management across the entire organisation.

Track and validate solutions

As long as you also apply a consistent and structured approach to finding and implementing solutions, you will then gain the main benefit you are after – reducing the occurrence of major incidents.

To make this happen, problem management also needs to apply a structured approach to finding solutions, getting approval to implement and tracking the implementation to an agreed target due date.

The results

Figure 3: Monthly occurrence of major incidents

Figure 3 shows what successful problem management looks like when you have skilled problem managers using a structured approach to finding root cause and finding and implementing permanent solutions. When problems stop causing incidents, the rate goes down quite rapidly.

Download an extract from Michael G Hall’s book

Click here to download an extract from Michael G Hall’s book, Problem Management, an implementation guide for the real world, which explains how problem management differs from incident management and looks at the differences between reactive and proactive problem management.

Read more on Managing IT and business issues