EN 38: Meta skills

Some teams talk about improving, they want to get better at their craft, produce higher quality code, deliver great products… but they never do, or take an incredibly long time getting there. There are many factors that contribute to it, some inside the team, some outside.

As always, things are complex and nuanced. High work in progress can make a team go nowhere. Since there’s no improvement, there will be decay. Decay of the architecture, or the quality of the code, an increase in bugs, lack of cohesion… you name it.

After many conversations, everything stays the same, there’s no action, no change, they keep doing what they do, business as usual. They didn’t rise to the levels of their goals, they fell to the level of theirs systems.

An awareness of pain points and improvements and a “will” to improve are important, but they don’t get you anywhere if you don’t act. Having the what (X is a problem, we need to do Y) helps. Understanding the why is essential, it makes us focus on the principles, fundamentals, the core, but when you want to improve, learn, adapt or, in general, when you want to do, we have to think about the how.

We wish to learn, aim to improve, how are we going to do so? We need meta skills: how to learn to learn, how to improve on improving, and doing so continuously.

The Toyota Kata was a research done to understand the routines and thinking behind Toyota’s success and how to replicate it in other organisations. Toyota’s success hinged on a culture of continuous improvement, but copying what they did to the T didn’t work out well.

What Mike Rother and colleagues realised is that Toyota, more than following specific routines, wanted employees to use scientific thinking, or a meta skill for developing solutions.

To change our way of thinking or doing, we have to think differently and act differently (and deliberately), in a reinforcing loop. Behaving differently will reinforce our way of thinking, and our way of thinking will reinforce our behaviour. If you’ve read Atomic Habits, that’s the same idea.

On one hand, we have our meta skill, and on the other hand we have practice. In the Toyota Kata, practice is done via, you guessed it, katas, and katas come at the macro and micro levels.

In essence, the “macro” katas are the Improvement kata and the Coaching kata, one for the learner and the other for the coach.

The Improvement kata has four steps:

1. Understand the direction or challenge. Where do you want to go?

2. Grasp the current condition. Where are you, what’s the situation?

3. Come up with a target condition. Where do you want to be next?

4. Experiment towards the target condition.

In order to go from where we are to our destination, the challenge, we have to take the first step, and then the next one, and so on until we get there. We don’t know how to get from our current condition to the target condition (the next step). There can be many obstacles in the way, and by experimenting, we’ll uncover and work on them and more importantly, we’ll learn more.

The key in the step four is that we are experimenting, as in coming up with hypotheses, and ways to test them and know more. We don’t know the solution up front.

The Coaching Kata is the other side of the coin and asks five questions:

1. What is the target condition?

2. What is the actual condition?

3. What obstacles do you think are in the way?

4. Which is the next step?

5. How quickly can we go and see what we have learned from taking that step?

On top of the five questions, it also asks us to reflect on the last step:

1. What did you plan as the last step?

2. What did you expect?

3. What happened?

4. What did you learn?

On the micro level, we have the starter kata for each step of the macro katas, to practice them deliberately and frequently.

The Toyota Katas share the same idea as the PDCA cycle, created by Shewhard, and adjusted by Dr. W. Edwards Deming to PDSA. The PDSA cycle is a “systematic process for gaining valuable learning and knowledge for the continual improvement of a product, process, or service”. In fact, these two concepts not only share the same ideas, but there’s a deep connection between them. Dr. Deming was a great influence on the Japanese economic miracle post-World War II, as John Willis writes in the book Deming’s Journey to Profound Knowledge.

The PDSA cycle, Plan-Do-Study-Act goes as follows:

1. Plan: identify goal or purpose. What are we trying to accomplish?

2. Do: Implement the plan.

3. Study: what did we learn, what went wrong? Monitor the outcomes to test that the plan is progressing and discover areas of improvement.

4. Act: integrate the learning, adjust the goal, change methods or ways of working…

With a meta skill, such as scientific thinking, that allows us to have a process about how to do something via observation, experimentation, reflection and learning, we can start “doing” in a deliberate way.

This kind of concept is not only exclusive to the PDSA cycle or the Toyota Kata. I believe this is a general idea that probably has been around forever, but we keep forgetting or ignoring.

In terms of deliberate and continuous practice, we have the idea of dojos, spaces for immersive and experiential learning, that has been applied beyond martial arts into coding, testing or agile, where people get together and practice their skills.

For the stages of learning, the Toyota Kata describes three stages that a learner goes through: follow, fluency and detach. We see this idea in Japanese martial arts (Shuhari) and also in creative endeavours, where we start by copying, then understanding the principles and theory and finally breaking free of the guardrails, ready to learn on our own and based on our circumstances.

Many teams start by copying a framework or process and stop there. The idea was never to learn or improve.

In fact, I think the lack of this kind of meta skill about continuous improvement is why, even though I’m a big fan of reflection, many retrospective ceremonies fail. We think that just because we talk—at best—every two weeks about improving, it will be enough. It becomes so ritualistic that it looses the meaning, there’s no intention and more importantly, a meeting can’t replace our ways of thinking, doing and being.

If the team wants to improve or talks about it—and you also care about improving—, not only map the issues (the what), certainly avoid simplistic solutions (observe the system), but focus on making improvement a habit, a deliberate process.

Interesting links

• Seeing like a bank by Patrick McKenzie. Interesting article about how banks are organised and why they’re bad at customer service.

• How to Make the Case for Slowing Down to Speed Up. Another great article by John Cutler that touches on the different strategies teams default to when needing to slow down and how to make the case based on value. Related to today’s topic.

• The huge climate impact of the middle classes.

• An Interactive Intro to CRDTs. A CRDT is a data structure that allows data to be shared across computers and updated independently—think collaborative editing, for example. Jake Lazaroff’s article is a good intro to these data structures.