When companies look for a manager, they should look for experience, right?

Well, maybe not. INSEAD professors Kishore Sengupta and Luk Van Wassenhove say their research has revealed what they call the 'experience trap.'

Conventional wisdom holds that as we do more things more often, we learn from experience and get better and better, and what we found in our research was that actually some of it may not be the case, Sengupta says.

Sengupta is an Associate Professor of Information Systems and Van Wassenhove is Professor of Operations Management and Director of the INSEAD Social Innovation Centre. What they found was that in big complex projects managers tend to draw on experience that doesn't help, or actually hurts the project.

In their research, they found that experienced managers applied the wrong lessons, didn't link the right lessons with outcomes or were given erroneous objectives or unhelpful feedback. It all adds up to extra problems in project management.

Managers tend to learn the wrong lessons and apply them again and again in a way that's fundamentally counter-productive, Sengupta says. Sengupta and Van Wassenhove used an intriguing simulation that mimics a real project development environment. As managers go through the simulation they have a number of decisions to make.

Many managers ran through the simulation and made errors, then they got feedback and tried the simulation again and made the same mistakes.

These things are deeply engrained into their experience, Van Wassenhove says. You have to help them unlearn what they've learnt and give them tools so they can avoid these types of mistakes.

A common mistake is in starting a project with one goal or one budget; and then when some of the variables change, as they always do in projects, managers decide to add staff but they fail to take into account time lags in hiring and training people, and projects end up late and over-budget. This is particularly true in software development, but Sengupta and Van Wassenhove say the lessons are widely applicable.

Most projects are complex in that the decision you take now, you see the effect later, Van Wassenhove says. Believing that experience will help you and then trusting that experience blindly without questioning the knowledge in the new situation is very tricky.

It's not that managers aren't smart, but projects are very complex, so managers make a lot of decisions over time and often for many projects at the same time. That makes it difficult for them to see the lessons they should be learning and to apply them to future projects.

If you are running 20 projects and you're dealing with a couple of hundred people and all these projects are in different stages of completion, it's very easy to fix something here and something there, and you lose track of action outcomes, Van Wassenhove says.

Estimates, goals and incentives are often unrealistic at the start of a project and as projects progress, things change and become more complex. Often the project managers actually know the goals and estimates are unreliable from the outset. But business culture dictates that managers shouldn't renegotiate or readjust, even when they have new information.

It creates a certain dynamic which leads to projects becoming more and more unproductive, Sengupta says.

Instead, companies should look at goals that induce the behavior they want to foster and then constantly review the goals to make sure they are creating the right behavior and outcomes.

And managers don't get the right feedback either. They get lots of reports and information, but it's not always very insightful and doesn't connect actions with outcomes.

Sengupta says it's not as though companies don't have the information or they don't want to give it to managers, but they don't give it in a way that the managers can understand what's going on and draw conclusions that can help them in this and future projects.