Optimism Destroys Projects

In popular TV programmes that follow home renovation projects, the projects never seem to run according to the initial schedule nor are they completed within the proposed budget. It’s a predictable end sequence.

  • Presenter: “What was your original budget?”
  • Programme participant: “X thousand.”
  • Presenter: “And what was the final bill?”
  • Programme participant: “I admit it did increase, in the end it cost Y thousand.”
  • Presenter: “And how long did it take?”
  • Programme participant: “We thought we would be in by X month but because of the weather/problems with a contractor [insert your choice of reason here] we ended up moving in Y months later than we planned.”

Writing about the reasons why major projects fail, Emeritus Professor Bent Flyvbjerg, a global expert in major programme management was unable to provide statistically significant data on how many major projects were on time, in budget and delivered the benefits anticipated. This was because the number was so incredibly small when set against the vast number of major projects that were delivered late (or not at all) and over budget. Whether it is the Channel Tunnel between the UK and France, numerous Olympic Games, the Three Gorges Dam in China, the price paid has, invariably, greatly exceeded the promised benefits. Major projects would seem to be a triumph of hope over experience.

Many amateur athletes set out to run a marathon and fail to complete it within the time they thought possible when they started training. You may well have been party to the conversation that goes something like this:

  • You: “How did you get on in your marathon?”
  • Participant: “I was hoping to run it in X but in the end it took a bit longer.”
  • You: “What happened?”
  • Participant: [Insert reason to suit, e.g…] The weather was hotter than anticipated/I was held up by the pack/I felt sick at mile 18.

Whether it is a mega project, a domestic project or an individual challenge, why is it things never seem to go according to plan?

Professor Flyvbjerg noted that attempting to understand the myriad failures in terms of changes in scope or complexity is to focus on the wrong thing. The biggest causes of the failure are behavioural. People are over optimistic about what they think can be done, they overstate the benefits and understate what the project will take to achieve, whilst ignoring (or selectively choosing) historical evidence of similar projects that reinforces their belief as to what can be achieved.

In the early 1990s, Roger Buehler, Dale Griffin, and Michael Ross asked students to estimate the time it would take them to complete the thesis for their degree. They asked the students for their estimate of both the shortest amount of time it could take and their worst case scenario of how long it could take. Their estimates on average ranged from 33.9 days for the best case and 48.6 days for their worst case scenario. The actual number of days was 55.5.

Our inherent optimism is essential for survival. It could be argued that we have evolved to be optimistic. We would never take on any seemingly impossible challenges unless we were optimistic. A lack of optimism is known to contribute to mental health issues. It’s easy to see how benefits get overstated as we need to really believe in the benefits we are going to get if we are going to expend significant effort, otherwise, why bother?

It seems that most major projects fail to analyse relevant historical data. For example, analysis of major IT projects paints a picture of many being over budget, delayed and ineffectively implemented. How come people are blind to what has gone before? Our lack of consideration of historical factors can be explained because we often see our approach and circumstances as unique and different to what has gone before, and so we focus on our plans and our optimism that we can achieve what we have set out to do. But as the saying goes, “Each of us is unique - just like everybody else.” There are exceptionally few things that are genuinely unique in totality. A project can be broken down into a series of steps, many of which are decidedly common. This commonality enables easy comparison with similar steps taken elsewhere to get a true sense of the time taken to complete and the likely cost in effort and resources.

In Buehler’s research, when the students were asked to consider their previous performance and that of others, their estimates of the time taken to complete an assignment became far more accurate. More accurate than the student’s estimates, were the assessment by others of their performance when armed with the same information. It seems we are much better at accurately predicting the performance of others than of ourselves.

How do we avoid the abject failure to deliver on time and on budget for so many projects? Two methods you may find useful, whether you are preparing to run a marathon, renovate your house, or implement a new ERP system or a major infrastructure project, are:

1.Examine how long similar projects have taken to complete.

You might think your project is genuinely unique. It probably isn’t. There will be something out there that is sufficiently similar that it will help you to inform your project.

For example, many companies have implemented new ERP systems such as SAP. There is a wealth of information on how long these implementations can take, the common issues that can create delays or problems, and the actual payback versus estimated payback.

There are websites that offer ballpark estimates of what specific house renovations can cost. This information can be used to validate proposed costs.

There is a vast amount of data on the average time taken to run a marathon based on age and time spent training. Whilst your circumstances will be a little different to others, by taking these ballpark estimates and adjusting them to take into account your circumstances you can arrive at a much more meaningful and accurate estimate.

1.Consider the skills and capabilities you have. How do these compare to the people who have completed similar projects?

Do you have the knowledge, time and resources that mean you are likely to do significantly better or is it the case that you are comparable to other people who have completed similar projects? If you have colleagues who have successfully implemented an ERP in another business then you may have more confidence than if you don’t.

If you are a site manager for a building company, it is likely you will be able to manage renovating your home with fewer issues than someone who has never managed a renovation before and has no technical experience.

An experienced endurance athlete is likely to have a better idea of how possible it is to train the required amount of time to achieve a desired finishing time in a marathon than a novice runner attempting their first marathon.

Just when you might be thinking, “Well, that’s all common sense”, the way we assess data can also be problematic. Take a moment to consider the research of Nobel prize winner Daniel Kahneman and Amos Yversky. They demonstrated that people will place an exaggerated faith in the stability of results from small samples. In other words, you may read one example of something going well and place a disproportionate value on that insight, and this influences the robustness of the decisions taken. The work of Kahneman and Yversky, Flyvbjerg and Buehler reinforces the point that it is not changes in the scope or complexity of projects that is the issue, it is how people perceive these changes and interpret the data that can lead to a bias to underestimate their impact.

It’s never fun correcting projects that have failed. But failure isn’t an inevitability. What it does require is a brutally honest assessment at the outset based on a deep understanding of base-line data. We need to temper our optimism with brutal facts. The trouble is, this is unlikely to make for great TV.

Dr Dominic Irvine & Professor Emeritus Simon Jobson