Playbook: Premortem
Here's a thing I like to do for any project of reasonable complexity, or length (rule of thumb: 2-4 weeks or longer). Pretend the project is done and it went really badly: What went wrong?
Getting a team together in a room and playing this premortem game is a surprisingly effective way of developing a shared understanding of a project, especially with respect to risks and how they might be mitigated or how contingency plans might be developed. Somehow inhabiting the perspective of our future selves can let us gain the insight of a retrospective before things go awry. In addition, the practice makes it easier for people to voice their concerns and discomfort with a project by providing a safe environment in which to do so.
Method
- Arrange a meeting, probably about an hour long, everyone should be involved.
- Before the meeting people should think of a couple of scenarios in which things go pear shaped. (Sent all your subscriber email addresses to a public source control system? Published an embargoed story too early, potentially causing the collapse of a major terrorism trial? Redirected all your web traffic to a competitor for a couple of hours? Sent an unencrypted list of blacklisted websites and swear words to every user of your app?1).
- During the meeting the group should work together to flesh out each scenario, work out how it might have come to pass and what should be done next time (actually this time) to avoid that outcome.
- Once have a list of how bad things might happen you can spend the end of the meeting figuring out how the new information fits in with the existing project plan.
That's it.
This isn't my original idea but I've been doing long enough that I can't recall where I first heard about it; I first used it whilst at the FT as tool for our election coverage planning probably having read about the practice in this HBR article.
Notes
- These are all made up and certainly not things that I have seen happen over the years