Premortem: Dissent Before the Plan Hardens
Featured image: ThisIsEngineering via Pexels.
A project plan is most useful while it is still cheap to change.
That is also the moment when the people around the table often know the most about how it could fail.
An engineer may know that a technical assumption is fragile. Someone from operations may know that a dependency routinely takes twice as long as the schedule allows. A person closer to the customer may already suspect that an important requirement has been misunderstood.
The point of planning is partly to get that knowledge into the plan before the work begins.
But there is a problem.
Once a team has spent time building a plan, raising those concerns can start to feel like opposition to it.
A team member may need to say that an executive’s assumption is unrealistic, that a favored technical approach looks fragile, that another department probably will not deliver what it promised, or that a deadline everyone has already committed to seems impossible.
Those observations may improve the plan.
Socially, however, they can sound like criticism of the people who created it.
Cognitive psychologist Gary Klein developed a technique called a premortem that changes the conversation.
Instead of asking what might go wrong, the leader tells the team to assume that the project has already failed.
It is six months or a year later. The project was a disaster.
What happened?
Team members first write down possible causes independently. Then the group collects them.
The obvious purpose is risk identification. Thinking backward from an assumed failure can make different possibilities easier to imagine.
But Klein also pointed to another effect:
A premortem makes dissent part of the assignment.
The person who says, “The customer never actually adopted it,” is not being negative.
The person who writes, “Our technical architecture couldn’t handle the load,” is not attacking the architect.
Everyone has been explicitly asked to explain why the imaginary failure happened.
That changes the social cost of raising an uncomfortable possibility.
The independent writing matters too.
If the most senior person speaks first and says, “I think our biggest risk is the schedule,” the rest of the room now has a reference point. People may start evaluating schedule risks rather than independently considering what worries them most.
Writing first gives each person a chance to form an answer before the group begins influencing it.
Healthcare safety programs have adopted the same basic structure. The Agency for Healthcare Research and Quality provides a premortem tool in which a team imagines that a safety-improvement effort has failed, writes down causes independently, and then shares them with the group.
The interesting lesson is broader than the premortem itself.
Organizations often say they want people to “speak up” and “challenge assumptions.”
But those instructions still rely on an individual deciding that an uncomfortable observation is worth the social risk of saying it.
A premortem takes a different approach.
It changes the structure of the conversation so that saying something unwelcome is no longer an act of opposition.
It is simply the job everyone in the room has been asked to do.
Sources and further reading
Gary Klein, Performing a Project Premortem
Join the discussion