Supporting reviews and performance: the role of the PMO

The SWWE Branch was delighted to be able to invite Graham Oakes back to the branch. Graham had last spoken January 2013 about how projects fail when they lose touch with reality. 

Tonight, Graham looked at the value of project reviews and assurance and how they can bring that much needed reality check, and how the PMO can help establish an effective assurance programme.

Graham asked why do projects continue to fail, and why are the causes of failure so consistent? Graham argued that as humans we are biased to ignore the actual situation and so to lose touch with reality. He referred to Daniel Kahneman's book Thinking Fast and Slow which explains that human biases tend to mask the actual truth of the situation. Thinking fast relates to our inherited survival instincts, where as thinking slowly is about analytical problem solving. Too often, when under pressure, we rely on fast thinking to make decisions. Failure only becomes obvious when we run back into reality. Reality cannot be avoided, it always gets you in the end, and by then it is usually too late to do anything about it. 

The challenge is how to find out about problems early enough to be able to take effective corrective action.

Graham argued that failure is OK, for example only 1 in 10 oil exploration wells are successful. The oil industry business models account for this and 1 in 10 is accepted as success in a high risk, high reward industry.

Graham discussed two types of failure: Inherent failure, such as the oil industry where failure is expected, and unnecessary failure such as poor communication, poor control, poor estimates, etc. The two should not be confused. Graham also made the valuable point that domain knowledge is essential for inherent failure, but generic project management skills are needed to avoid unnecessary failures.

Good quality information is essential to provide warning of failures. Statistical information does not normally help the project manager as they are usually unable to manage issues outside the scope of the project. However portfolio and programme managers can use statistical information to make decisions about prioritising resources across portfolios and programmes to meet strategic goals. However they often do not get suitable information; reports are too detailed, and often tainted with project manager optimism bias. Organisational culture and behaviours can discourage honesty: bring me solutions, not problems! Also project managers often do not have time to report properly, reporting is often done as an afterthought late on a Friday evening!

The PMO can help provide quality information, cutting through the biases, avoiding the 'thinking fast' survival instinct, and focussing on the slow thinking analytical information. The PMO can help provide reliable assured information at a number of levels; at the strategic level for external communications and governance; for portfolio and programme decision makers; and information to help project managers to understand the reality of their projects. 

Graham outlined a number of key considerations for setting up an effective review and assurance system:

• Make reviews part of normal business, not an extra to be done if a project is in trouble. This creates the right behaviours, with the project managers preparing and thinking properly and increases the chance of them finding potential failures early themselves and dealing with issues in time.
• Be clear about the customer for the review information. What level of information do they need to be able to make the necessary decisions?
• Assurance should be risk based. The cost of assurance should be commensurate with the risk of the project to the organisation. Typically 1 -2 % of total project cost would be reasonable.
• Use an appropriate mix of review types. Large public sector projects use expensive OGC gate way reviews. However, lighter-touch regular reviews; monthly or weekly can be very effective to encourage the right behaviours as can an informal ‘buddying’ approach with a trusted colleague over a coffee.
• Run each review as a project with clear objectives.
• Run reviews as a structured process. Train the review staff to use the process, tools, techniques, interview skills, how to stand back and be objective.
• Focus on actual evidence. Look for collaboration of interview opinion from documentary evidence and triangulation. Real evidence is needed to cancel zombie projects!
• Manage the logistics of the review: access, security, documents, interview schedule.
• Minimise the overheads on the project team. Use standard reports, don't ask for special ones!
• Act on the findings! Otherwise you are wasting your time. Ensure actions are tracked and monitored.
• Evaluate the effectiveness of the review and assurance process. Does it meet stakeholder needs for information? 
• Make full use of the lessons learned from reviews across the organisation's projects. Reviews are a useful way to train and provide experience for less experienced PMs.

In summary Graham concluded that assurance provides an independent reporting line in support of the project, it does not replace normal reporting lines by the PM.

Project review and assurance is an invaluable tool to provide regular, independent, checks of reality.

Martin Gosden
SWWE branch chairman