Predictable improvements with unpredictable humans. Imprecise and reliably better.

Is this predictable?

Water boils differently on a mountain than at sea level. Same water, different context, different result.
Humans are no different: complex, yes but reliably influenced by the context we’re in.

I talk to leaders who want a culture of learning, innovation, resilience, psychological safety, and operational excellence. “Can you really design culture?” they ask. “Humans are too complicated!”

Humans are complicated. And directionally predictable based on the context. And when you design the context - physical, social, and workflow - you tilt the odds in your favor. Reliably so.

Not a 100% guarantee with dates and precise numbers (only gravity gets close to that, and even gravity changes off-planet). But better, predictably so - much better.

1) Workflow: protect attention

In one team, Slack pings, email pop-ups, and “got-a-minute?” interruptions are the norm. In another, they batch messages, protect maker time, and run “no-meeting mornings.” Which ships more meaningful work?

We know interruptions are costly. After a disruption, people can take over 23 minutes to fully regain focus. That’s not a character flaw; that’s a human brain doing what human brains do. Design the day to reduce needless switches and you recover a huge chunk of your productivity, predictably so. (It's estimated that companies lose up to 94% of productivity due to interruptions)

Try this: Make 9–11am a company-wide deep-work block. Batch notifications. Close the door on “just one quick thing.”

2) Communication: make the important thing the default

The WHO didn’t ask surgeons to “try harder.” They introduced a Surgical Safety Checklist - a simple ritual. Results: in a multinational study, complications fell from 11% to 7% and deaths from 1.5% to 0.8% after adoption. That’s design, not heroics.

Try this: Build micro-checklists for handoffs and recurring decisions. Turn good decision-making process into a default VS hoping for better humans.

3) Social norms: build psychological safety

Google’s Project Aristotle found that the #1 factor in team effectiveness wasn’t IQ, tenure, or passion - it was psychological safety: people feeling safe to speak up, ask for help, and surface risks. This isn’t a vibe; it’s a trainable norm. Amy Edmondson’s research shows psychological safety enables learning behaviors that drive performance (various studies citing quantifiable benefits such as a 50% increase in productivity, up to a 76% increase in engagement, and a 27% reduction in employee turnover). 

Try this: Start performance review with “What did we learn?” not "Why did it go wrong?" - Rituals change how we feel, feelings change how we show up.

Not scientific AND it works every time

“But coaching and culture aren’t exact sciences.”

True. And yet in my 18-year coaching career I'm yet to see a case when over time we didn't exceed our set goals. However complex.

People are complex AND context-adaptable. And context can be designed for. Over time, when you apply tested methods from behavioral science, psychology, coaching benchmarking, communication research - attention design, checklists and protocols, psychological safety rituals, and smart defaults - results are predictably better.

Even with complex humans.

Over to you dear reader, when you’re trying to change your own behavior - or influence a team - are you relying on hope or are you deliberately shaping the environment, the workflow, the norms, the processes and the defaults?

A 2-week experiment (steal this)

  1. Pick one outcome you want 20% more of (fewer defects, more results shipped).

  2. Choose one lever above.

  3. Install a tiny ritual or default.

  4. Track the metric for 10 working days.

  5. See how the needle moves forward. Predictably, with unpredictable humans.