What's the lean take on learning?
Dear Gemba Coach,
Our previous CEO used to preach “servant leadership,” but now’s he’s left and the new style is far more traditional command and control. I’m personally interested in learning leadership and wondered what the lean take on it would be?
Interesting question – I have to confess I’m not well versed in servant leadership. As I understand it, this is a reaction to autocratic leadership, putting forward a style of leadership based on convincing, encouraging, and supporting subordinates in achieving their goals. As I understand it, the emphasis of servant leadership is on people’s growth and well-being, sharing of power, and putting the needs of others first. I can well understand this is not really in the spirit of the times, where authoritarian leaders seem to be back in fashion, with a vengeance.
I’m not quite sure what you mean by “learning leadership” – I’ll assume this is the form of leadership prescribed by learning organization enthusiasts. In this sense, a learning leader would need to master the five disciplines of the learning organization:
- Be a system thinker: Understand that the whole is more than the sum of its component parts, and that dynamic behavior is created by the effect of feedback loops on key resources (stocks) and their consumption (flows).
- Commit to personal mastery: be personally explicitly committed to a process of learning, both for oneself and others in the organization.
- Work with mental models: Mental models are seen as the basic representations through which humans see the world and react to events. Working with mental models means shifting people’s understanding of the situation so they themselves see what to do rather than just telling them to do something. This involves a triple-learning process of (1) learning to do something new, (2) as a result of this learning effort, learning to change the assumptions held about the situation and discard wrong beliefs, and (3) learning to change the values underlying these assumptions and discard non-adaptive values.
- Develop a shared vision: A shared vision creates a common identity and is a source of motivation if it is indeed shared, and not developed by the top and imposed on others, but built on the individual visions of employees at all levels.
- Facilitate team learning: Encouraging sharing of learning across individuals in teams, and across teams in the organization to disseminate learning faster. This involves creating organizational structures that support exchanges and communication and encourage individuals to engage in discussion and dialogue.
I would certainly agree to the intentions of both servant leadership and leading from practicing the five disciplines of the learning organization, but I have to confess that, from the Gemba, I find such abstract discussions not that helpful.
At a more hands-on level I find “learning” poses a few very practical problems.
Learning in a Lean Context
First, although behavior changes all the time, not all changes are learning. Learning means a change that improves performance to reach a goal – it’s quite specific. In this sense, simply adapting to the pressures of a situation is not learning. True, this is a grey area, as not all learning is deliberate and sometimes, you “learn” by repeatedly reacting and sometimes this evolution is positive (better be lucky than good, any time) – but I prefer to restrict “learning” to a deliberate activity of clarifying a goal, narrowing down the new way we want to learn, and practicing it until we get it right.
Secondly, not all learning transfers from one team to another. No two situations are really alike, and no two teams are alike in terms of their balance of skills and characters. As a result, the “learning” one team achieves – that is the new routines it learns to perform better – might not necessarily apply to another team. I’m not suggesting the wheel needs to be reinvented from scratch every time, and certainly a good idea from another team is a great place to start, but you can’t bypass the learning process either in the belief that solutions apply cookie-cutter style.
Thirdly, most general ideas about learning (and perhaps leadership?) are likely to overfit. In statistics, an overfitting model is one that is over-complicated for the data set, and thus describes noise rather than signal. When discussing concepts, I tend to think as overfitting any explanation that is too vague to distinguish noise from signal. For instance, most leaders would consider themselves to be “learning leaders”, whether they are or aren't, because they will cherry-pick instances in which they have learned something and in which they have taught something to someone else – even if, in the main, they are, as you say, traditional command-and-control types in their everyday interactions.
How can we make the “learning leader” idea operational? Let’s go back to the Gemba and visualize three clear cut events that describe learning in a lean context:
First, kaizen. I was on the Gemba last week with a hospital CEO and we could clearly distinguish the wards in which the team was tackling a specific problem to improve the quality of its care – or not. Some teams were deliberately trying to learn something, such as getting patients more quickly to X-rays and back, or reducing bed-sore bedridden patients and so on. Many teams were just trying to get by one day after the other without major incident and not trying to learn anything specific other than just doing their job. It’s a pretty clear-cut difference. The challenge for the CEO was (1) how to support the teams that did try – and encountered all sorts of obstacles – so that they kept at it, and having learned one thing, move on to the next issue; (2) convince teams that weren’t trying that they could indeed, try to face a problem to learn one thing.
Second, acting upon the kaizen. As we listened to the teams that had done a kaizen effort present what they’d learned, the CEO would ask the department head, often a physician, what procedure needed to be changed in order to learn from what the team had learned. In most cases, this question led to aggrieved silence, and then a debate about how nothing could be changed in the hospital in the first place – an issue for the CEO himself. There, we were faced with the second clear-cut learning event that needs to happen. Having led learning of others, the leader must now demonstrate learning for themselves, by changing a procedure, rule, belief – anything – as a response to the learning demonstrated by the team on the Gemba (what we used to call in the previous century “double-loop” learning. This is not a small step.
Thirdly, a pull system acts as a learning compass. A specific lean question is: what next? If you are really trying to create a learning culture and you are the CEO of a 3,000 people hospital, clearly you can’t okay every learning effort personally. People have to choose their own kaizen topics autonomously. The question, therefore, is: how to they choose what to change when you’re not there? This question needs to be taken very seriously because it is the undoing of many “learning” approaches, from six sigma to holacracy. It’s great that people choose learning projects, or change what they do autonomously – but how do you know they’re going for the right changes. The answer is you don’t, and when they do a couple of wrong changes with negative impacts on patients or effectiveness, they have to be told to back down, which creates a whole drama in itself and mostly stops dead any further initiative. The lean trick is a pull system: by constantly reducing lead-times and clarifying flows, the “rocks” appear in sequence, so the problems to tackle are pretty visible. In the hospital context, the first step of the pull system are highlighted by (1) a “patient promise” board -- when did we tell the patient they would go home and what happened if we can’t keep that promise. (2) Work on patient flows and streams and (3) putting logistics at the service of the wards as opposed to the other way around (as it is in most hospitals I know). The pull system visibly points towards the right problems to solve.
Learning, like any dynamic behavior, is hard to pinpoint. In the case of hospital wards, learning has three clear components:
- Making sure that we know what we know by checking one standard a day – the local manager picks a problem from the day before, explains the standard and asks every nurse to check how they perform this by themselves.
- Improving our ways of working one by one – using a structured kaizen approach to study our own working methods and focus on one improvement point after the other.
- Taking greater responsibility for our Work environment – and understanding better how the physical Work environment affects Work. For instance, in the hospital care, “5S” is foremost about mastering aseptic techniques.
The leader's job is, therefore, first to make sure that all wards engage in this hands-on learning – which is quite a challenge in itself. But then, the second challenge is for the leader himself to engage in learning personally (and with the department heads) to change the hospital procedures according to what has been learned. Finally, and this is unique to lean, the leader must also create the pull system that will act as a compass and orient every team towards true learning.
Are You Narrowing Your Problems Down?
"Rationality did not lay in higher reasoning powers, in visionary schemes, but in the ability to narrow down problems until one reached the nitty-gritty level at which one could actually do something about them," writes the protagonist of Michael Balle's The Gold Mine.
Lead With Respect Shares Tangible Practices That Develop Others, Says Author Michael Balle
Michael and Freddy Balle's book Lead With Respect portrays on-the-job behaviors of lean leaders which can be learned through practice. Michael explains how these can help fulfill the promise of lean by aligning the company’s success to individual fulfillment.
How Can Lean Affect Shareholder Value?
Lean can help challenge assumptions and surface opinions that ultimately improve shareholder value, argues Michael Balle.