Give me a little chaos any day

There is something reassuring about a plan that has worked before. We know the steps. We understand the roles. We can point to a result and say, with some confidence, “Let’s do that again.”

As a project manager, I can see the appeal. People need to coordinate their work, make commitments and know what others are relying on them to deliver. Predictability makes that possible.

But success can teach us a little less than we think.

We know what we did. We know what happened afterwards. Understanding exactly why it worked is harder. Perhaps the process was excellent. Perhaps a few people quietly compensated for its weaknesses. Perhaps the timing was favourable, or the team had enough spare capacity to absorb the things we hadn’t anticipated.

Repeat the process with different people, less capacity or a change in the outside world, and the result may be quite different. Yet we can find ourselves looking for a failure of execution before we question the expectation.

I’ve been listening to J. Doyne Farmer’s Making Sense of Chaos. Early in the book, he distinguishes between models built on assumptions about how the world works and statistical methods that use regularities in data to make predictions. He also warns that those regularities can shift when circumstances change. That got me thinking about how we manage organisations.

“It worked before; it will work again” sounds like evidence informing a decision. It can also conceal a substantial assumption: that the conditions which made it work are still present.

The people will respond in the same way. The incentives still make sense. Other teams have the capacity we need. Our customers want what they wanted last time. Nothing significant has changed beyond the boundaries of our plan.

We rarely say all of that aloud. We put a date in a box.

Over time, the expectation can become something people are required to defend. Reporting begins to revolve around whether we are on track, with less attention given to whether the track still leads somewhere useful.

Which brings me to the part where I explain why leaders need to embrace a little chaos, offer three practical suggestions and finish with a sentence you might consider sharing.

You could probably see it coming.

Hang on, though. You might also have spotted a problem with my argument.

A prediction isn’t an assumption.

You’d be right. If I expect a team to complete a task by Friday, that prediction might be supported by experience, an understanding of the work and a sensible conversation with the people doing it. Calling it an assumption would flatten a useful distinction.

Equally, you’ve been reading this on the assumption that I’ll take your attention somewhere worthwhile. You may have predicted the shape of the article. Those are related, but they aren’t identical. And if I start throwing unrelated words onto the page, I haven’t demonstrated good leadership. I’ve wasted your time.

So I need to be more careful about what I mean by chaos.

And I need to let your possible objection change where this post goes.

Predictions depend on assumptions. Some are well supported; others are little more than habit. What concerns me is how easily we lose sight of the difference, especially when a familiar approach has the weight of previous success behind it.

Once that happens, an unexpected outcome can feel like a challenge to our competence. A colleague questioning the plan can sound unhelpful. Someone doing things differently can appear to be introducing unnecessary risk.

I wonder how much useful information gets lost in those moments.

Consider a team that repeatedly meets its deadlines because experienced colleagues work around a failing process. From a distance, the results look dependable. The prediction that next month’s work will arrive on time seems reasonable. But the people delivering it may know that the arrangement is becoming unsustainable.

If our response to their concerns is to insist on the established approach, we may preserve the appearance of control while weakening our ability to deliver.

At Rocket, we used a phrase as a signal - “That’s the way it’s always been” said with withering effect and meaning we’ve accepted past patterns as good enough.

The little chaos I want makes room for that conversation before the arrangement fails. It allows someone to interrupt the reassuring account of how well things are going. It gives a team permission to try a different approach on a scale where we can learn from the result. It leaves room for an answer the manager wasn’t expecting.

That asks something of leaders, too. We need to make commitments while accepting that our understanding is incomplete. We need to distinguish a dependable standard from a familiar habit. And we need to notice when our desire for reassurance is making it harder for people to tell us what is happening.

None of this gives us permission to be careless with other people’s work. An experiment needs boundaries. A change of direction needs an explanation. People deserve to know what they can rely on, particularly when so much else is uncertain.

But they also deserve to work somewhere that can respond when reality refuses to follow the plan.

I still want to know what worked last time. I’d like to know what made it possible, what it cost the people involved, and what has changed since.

And I want someone in the room who feels able to say, “I don’t think that will work this time.”

Even if they interrupt me.

The book that prompted this post: J. Doyne Farmer, Making Sense of Chaos: A Better Economics for a Better World, Introduction. Publisher’s sample.