Skip to content

Managing biases that affect project decision-making

Added to your CPD log

View or edit this activity in your CPD log.

Go to My CPD
Only APM members have access to CPD features Become a member Already added to CPD log

View or edit this activity in your CPD log.

Go to My CPD
Added to your Saved Content Go to my Saved Content
Shutterstock 648098419

Biases are evident in all walks of life. They may be deliberate or subconscious, but however they materialise, they affect your decisions.

There are over 180 psychological biases that humans – including project professionals – are prone to. For example, optimism bias leading to planning fallacy – the tendency to overstate the forecasted benefits of a project and to understate the timescales and costs.

“It’s always going to be a ‘good project’,” says Pat Weaver, director of Mosaic Project Services, based in Melbourne, Australia. “The whole industry seems almost incapable of taking a more pragmatic view. Projects invariably run late but, ‘Oh no’, the project manager says, ‘this one will be different.’”

Anchoring bias is another example – where project professionals become anchored to early cost and time estimates, even if they are not always correct.

Confirmation bias, meanwhile, is where project professionals value new evidence that supports their strategy and reject evidence showing a potentially different view. It is linked to the illusion of control, whereby project professionals overestimate their ability to control events.

Age and experience don’t necessarily help

Weaver recounts a project in which a group of civil engineers were chosen to construct a wind farm.

“They believed in their traditional culture and methods and did not access the knowledge of turbine manufacturing experts,” he says. “They were too optimistic and insular. Millions of dollars were wasted.”

According to a 2019 study by Alexander Budzier of Saïd Business School, only 0.5 per cent of nearly 12,000 projects were delivered on budget, on time and with the desired benefits because of forecasting errors caused by bias.

Tackling bias is easier said than done. Even recognition and experience of bias don’t automatically change our mindsets.

Matthew Todd, director of rail at Lanes Group, offers an example: “As a director I have a more pessimistic bias than I did when I was a younger project manager. I think more about the whole business now. But I am currently doing a personal project building a swimming pool for a local children’s sports charity, and I find myself stressing wholeheartedly the community benefits. I want it to be the best in the south-east!”

Todd’s story recalls the 1975 book The Mythical Man-Month, by Fred Brooks.

“Brooks said that IT projects were coming in late because young people were too optimistic. But he believed that, as they got older, more realism would emerge,” Budzier states. “Well, 50 years on, the data shows it didn’t happen! Bias has been hardwired into our brains over centuries.”

So how can we manage bias – if at all?

Project professionals can use corrective uplifts to tackle optimism bias. These increase estimates of a project’s costs by as much as 30 per cent, and decrease and delay the receipt of estimated benefits.

This is linked to reference class forecasting (RCF), whereby project professionals identify a sample of between 20 and 30 similar past projects. They use this data to identify cost overruns and adjust their current estimates. It’s being used by the Department for Transport, among others.

RCF’s proponents say it has reduced cost overruns, with the benefits particularly visible in the latter half of projects. This is a time when, in most projects, project managers would be renegotiating scope and trying to raise new funding. However, under RCF, they simply get on with completing the job.

“RCF is an easy method, but requires organisational data maturity,” says Budzier. “Data is often not easy to access, and more organisations should improve their data management.”

Paul Adamson, partner and global head of delivery at PA Consulting, is unconvinced. “Past learnings are there, but biased people will wilfully tune down the negative side and tune up the positive side,” he says. Instead, he believes in the power of communication.

“Project managers need a safe space where they can talk to third-party stakeholders and identify biases,” he says. “Don’t listen to the nodding donkeys around you. Talk to those without skin in the game. ‘Is it time to change course?’ This does not happen often enough.”

Project professionals also need to build diverse teams, including dissenting voices to tone down biases and groupthink.

“It is only when you start talking to a much wider community that you hear other opinions. If you are flexible, it could change your preconceived ideas,” says Todd.

The impact of the pandemic on bias

Weaver says lessons can also be taken from iterative project management. “Yes, you have budgets and deliverables, but you don’t get too excited about planning infinite detail,” he explains. “In addition, it involves a lot of team communication and questioning.”

The pandemic could mean more answers are needed.

“Project managers will be preparing for the next catastrophe rather than looking at a project on its own merits,” Budzier warns. “It could lead to catastrophe bias.”

Remote working could also lead to more ‘stressed’ projects. “Optimism biases will be identified at a later stage because we’ve been working remotely,” Adamson says. “Project managers are not being challenged by others. They need to go that extra mile and ask more questions to ensure they and their teams are digging in against biases.”

Of course, some biases, such as optimism, can have positive effects too. If everyone was too cautious, then innovation would wither. However, finding much more space for realism in our forecasting will secure even greater progress.

Find out more about reference class forecasting by listening to The APM Podcast’s interview with Bent Flyvbjerg.

This podcast is hosted by a third party (Podbean.com), you cannot listen to this podcast unless you accept statistics cookies to be able to listen to this audio content.

You may also be interested in:

0 comments

Join the conversation!

Log in to post a comment, or create an account if you don't have one already.