Saturday, July 21, 2018

Why Post-Mortems Go Wrong (And How To Do It Right)

Post-mortem is a great idea. At the end of a long project, everyone sits around to openly share their thoughts about what worked and what didn't, and what to adopt and what to avoid in future. We can learn a lot from such constructive discussions. We can improve as individuals and as a team.

But more post-mortems are badly executed. Discussions spiral into heated debates. Everyone gets super defensive of their own turf. The exact causes of things that went wrong aren't properly identified. The right lessons don't get learnt. The bad lessons turn into policies.

Whether it's at office with senior managers and co-workers, or at college with administrators and students - post-mortem is almost always a total waste of time.

Why is this so?

Post-mortem rocks!!!

1. Passing the Blame

Everyone has an ego. Everyone has vested interests. Everyone has an agenda.

People will say things like "If we were given more time to prepare, we could've done a better job."

No shit, Sherlock. Thanks for stating the obvious. More resources would usually get better results (although, in some cases, this is debatable as well). But your task was to attain Goal X with Y amount of resources. And you failed. Belatedly complaining about how Goal X is unrealistic or Resources Y wasn't enough just smacks of being a sore loser.

Anyway, what's glaring is not so much the merits behind their criticism (unrealistic goals and lack of resources are still valid points), but their motivation.

The mistakes were done way before they got on board. Their great plans were wrecked by bad luck. They could've done better with more support.

It's always someone else's fault that things went wrong.

They'll almost never say, "I spent so much time focusing on Task A instead of Task B - my bad".

Or: "I sucked. I couldn't handle the pressure. I need to seriously rethink my role in the organisation." 

Post mortem should focus more on internal self-reflection, rather than external finger-pointing.

2. Passing the Work

Some post-mortems are actually quite positive.

"Next time, we should be doing this and that."

Sounds great, right?

Except that there is no next time, because the 'we' here refers to the 'other people'.

I once attended a post-accident meeting. Serious stuff. A lot of people got hurt, it was all over the papers. The best brains from different departments huddled to come up with a plan on how to avoid such tragedies in future.

Instead of sticking to their turf, people started poking their noses in other places. Legal being critical on Engineering stuff, Engineering being critical on Marketing matters... Quite a comedy skit.

Whilst thinking-out-of-the-box interdisciplinary creative thinking is to be encouraged, most suggestions were just practically unworkable, if not theoretically flawed. Worse still, people were just throwing out random things that came to mind, without doing prior homework.

Consider the following exchange:

Legal: "Why didn't the valve shut off?"

Engineering: "The fail-safe mechanism, well, failed."

Legal: "Shouldn't we have a back-up system if the fail-safe fails?"

Engineering: "The workers were supposed to monitor the operations, and intervene if the fail-safe failed for whatever reason."

Legal: "And if the workers weren't paying attention - like in our case - then what?"

Engineering: "Then... accidents happen. Look, no system can be 100% accident-proof. We already employ the Swiss cheese risk-mitigation model in our design..."

Legal: "Design a second fail-safe!"

Engineering: "It's not our design. We got it from our suppliers."

Legal: "Then source a better supplier with a better fail-safe mechanism. You know, initiate a tender process..."

Engineering: "We already do. The design is pretty universal."

Legal: "You guys should design a better valve mechanism then!"

Engineering: "Sure, we'll put 'reinvent the wheel' on our action item list as well. Now, moving on..."

The passing-of-the-buck syndrome also applies >within departments. One team botches up a project, then makes a list of useless recommendations for the next team to pick up.

"Double the work hours! Have more regular meetings! Get more members on board!" 

Er... no thanks. Your work is bad, and so are your ideas. Right this way to the exit, please. No pushing, no talking...

Star Trek's latest super villain

How To Improve Your Post-Mortems

Here are some constructive tips on running post-mortems.

a. You can only comment on what you did wrong

Stare into the mirror, and reflect on where you screwed up. That's all. No finger-pointing. The best part is, everyone else will start to feel bad for you and point out things you did well to cheer you up. Positive vibes, yay! And even when people give you suggestions on how you can improve, they're coming from a genuine desire to help you (and not to put you down). It's a good constructive feedback loop.

b. You can only comment on matters not involving yourself if you have sufficient knowledge and expertise in the area

No more "we should've set up the day before and do more rehearsals" crap (Hello, the venue was only available few hours before the event, were you even listening during meetings?). Fact-check and think before you speak about someone else's responsibility. Don't speak for the sake of participating. It's better not to give any ideas than to give bad ideas (which may end up on the action item list).

c. You can only recommend major changes if you're executing them or accountable in their execution

Brainstorming ideas are easy-peasy. Execution, not so. So if you wish to bring up any 'bright' ideas, you better make sure you're also involved in the planning, execution, or at the very least, monitoring. Otherwise, just shut up. If you have no 'skin in the game', very likely your ideas suck. And it's really unfair to push your untested ideas for others to execute and take the blame if they don't work.

Post-Mortem About Post-Mortems

Yes, I know it's rather ironic. In essence, this post is a post-mortem about post-mortems.

But it's really critical to do your post-mortems right. Otherwise, you'll just keep on repeating the same mistakes over and over again.

Post-mortems are meant to find good answers, not raise stupid questions.

Post-mortems are meant for reflection, not deflection.

And above all, post-mortems are meant to get shit done, not throw them all around. 

No comments :

Post a Comment