OneOne Hundred Blog

Why we are not learning from Project Failures

With approximately 70% of all IT-enabled projects considered to be failures, a question that is regularly asked is "Why do projects fail?" There have been extensive studies on this subject, but perhaps a more pertinent question should be "How do we learn from project failure?" A good starting point for such learning would be to consider the lessons learnt from individual projects and to build this leaning into improvements in our project methodologies.

Lessons Learnt Reports

One of the standard PRINCE2™ documents is a "Lessons Learned Report", the purpose of which is to bring together any lessons learned during the project that can be usefully applied to other projects. It is intended that at the close of the project a Lessons Learned Report is compiled and disseminated. OGC guidance recommends that as a minimum, lessons learned should be captured at the end of each stage of the project. This is good advice, as project teams inevitably change during the lifecycle of the project and it can be difficult at the end of a long project to accurately recall how it was progressing during the early stages.

The PRINCE2™ product definition for a Lessons Learned Report suggests it should contain:

an examination of which management and quality processes: went well, went badly, and were lacking;

a description of any abnormal events causing deviations from plans;

an assessment of technical methods and tools used;

recommendations for future enhancement or modification of the project management method;

useful measurements on how much effort was required to create the various products; and

notes on effective and ineffective quality reviews and other tests, including reasons why they worked well or badly.  

The suggested content appears eminently sensible and by sharing this information project teams could significantly improve their delivery performance - highlighted by the fact that 70% of projects that are judged as being failures. Research reveals very few publicly available Lessons Learnt reports and the handful that are available do not appear to conform to the PRINCE2™ product description, despite this being the mandated project management standard throughout the UK public sector.

NAO Report - Helping Government Learn

In a report entitled "Helping Government Learn", the NAO refers to a finding in 2007 from the Committee of Public Accounts that "Over 2,500 Gateway Reviews have been completed to date, generating large amounts of valuable information. Little attempt has been made, either by OGC or departments, to mine the rich vein of knowledge they must have generated."

The NAO also discusses the role of the Programme and Project Management (PPM) Centres of Excellence (CoE), which have been existence for the last 5 years, and whose role is to strengthen the delivery of government IT-enabled programmes and projects.  NAO's found that over 50% of the departments surveyed rated their CoE as ‘effective', but only around 25% of CoE's currently produce a regular report of lessons learnt from programme and project delivery. The report also states that two-thirds of departments do not have a formal requirement in place to search for relevant knowledge, skills and experience at the outset of a project/programme, which is a requirement of the PRINCE2™ methodology.

This lack of a formal requirement to search for relevant knowledge, skills and experience is interesting as Information Note 5/2001 published by the OGC, states that:

"The lessons learned from the first 27 completed Reviews have been identified and break down into the following categories:

a lack of relevant skills and business resources;

a need for more clearly defined success criteria;

a need for better definition of roles and responsibilities;

a need for better risk management; and

a lack of good market knowledge and procurement advice."

So what is the barrier to learning the lessons

An article in the IET's Engineering and Technology magazine entitled "Agreeing to Fail" examines two potential causes of management failures, Groupthink and the Prospect Theory, which seem particularly relevant to project failure. Consider how they may apply to management's failure to learn lessons from project failure and its reluctance to insist on applying formal lessons learnt processes.

Groupthink

Groupthink is a type of thought exhibited by members of typically homogenous groups who try to minimize conflict and reach consensus without critically testing, analyzing, and evaluating ideas. During Groupthink, members of the group will avoid promoting viewpoints outside the comfort zone of consensus thinking, typically to avoid looking foolish or embarrassing other members of the group. In highly structured organisations, junior members of the group may be reluctant to raise issues, because senior members may perceive the ideas as foolish or a challenge to their authority, even through the more junior members may have the relevant and current practical experience, e.g. of dealing with service users or customers. From a project perspective Groupthink may initially result in a reluctance to challenge unrealistic estimates or timescales, and then after the event the avoidance or an unwillingness to consider why the estimates or timescales were unrealistic in the first place.

The Prospect Theory

Prospect Theory is an attempt to describe decisions between alternatives that involve risk, i.e. alternatives with uncertain outcomes, where the probabilities are known, and therefore takes into account the decision-maker's perception or risks and reward. It frames decisions in terms of two domains, those with gains and those with losses. In the case of the domains with losses, the choices are likely to lead to some degree of loss, individuals will tend to pick the risky option. Where groups are involved in making the decision, this will tend to result in a riskier option being chosen than if an individual manager had made the decision on their own. Research shows that where decisions have led to fiascos, the choice was framed in terms of two or more unattractive options. Research has revealed that organisational disasters can be foreseen, but that signals are not seen as warnings because they are seen as consistent with the organisational beliefs and aspirations. So the signals that there are problems with say the deployment of a new system will typically be ignored during the project and then afterwards simply rationalised as - we always have problems with a project of this type or with a project of this scale, rather than learning the lessons why.

Conclusion 

There appear to be reluctance to producing and disseminating project lessons learnt reports. From our experience managers are often uncomfortable with having the basis for their decisions questioned, and the presence of Groupthink or Prospect Theory could explain why projects can so readily pass through early warning signals.

In a future article we will review some of the published lessons learnt findings and consider whether they actually address the criteria listed in the Lessons Learned Report product definition above.