The Purpose of Gemba Walks
I recall interviewing executives in a conference room at Toyota’s proving grounds in Arizona when I was writing The Toyota Way. The President of the Toyota Technical Center, George Yamashina, on assignment from Japan, who had responsibility for engineering operations in North America, including the proving ground, stuck his head in the room. Immediately everyone stopped talking and turned to look at him. He smiled and said, “please continue on. I have work to do.” He then disappeared. He was on a Gemba Walk. He was there to “confirm the processes.” He needed no escort, and wandered everywhere, observing, asking questions, joking with people, but not giving advice. After he returned to Michigan his observations and advice came rolling into the senior managers—no direct orders.
Mr. Yamashina is a veteran at “Gemba Walks.” He does not need a coach shadowing him and he does not need written standard work to tell him where to go or what questions to ask. He has a very clear vision of what should be happening and compares that to what is happening. He is looking for gaps that need to be addressed. What should be happening is more than results, but various processes. For example, there should be processes for improvement in place. People at all levels should be able to clearly state their objectives for the year (coming out of Toyota’s annual Hoshin Kanri process). They should be able to share with Yamashina simple charts that show their actual performance versus targets and the gaps should be clearly highlighted with yellow or red on visual boards and succinctly describe what they are working on now toward those objectives.
What I just described is a perfect case scenario. But what is the value of a Gemba Walk if the leader doing the walking is a novice at it, there are few visuals to observe, and there are no clear objectives in place? This is the situation many organizations face. In this case, it is important to have an experienced coach walking with the leader. They should have a well-defined plan for the Gemba Walk. The coach should avoid intervening unless the leader does something disrespectful and save feedback for periods between different areas visited and for after the walk.
Here are some things that should not be happening on the Gemba Walk:
- Leader breaking from standard work to ask or comment on whatever comes to mind.
- Leader noticing things he does not like and demanding to know why they aren’t being fixed (e.g. Why are all those empty boxes sitting on the floor? Why is the machine down and who is coming to fix it?).
- Leader - walking with a Lean coach who is supposed to be giving feedback but is afraid to say anything critical to avoid getting in trouble.
I have been in many situations where I was asked for advice on how to conduct Gemba Walks. The conversation might go something like this:
Continuous Improvement Leader: We have started a program of leader standard work. We want all of our leaders to do daily Gemba Walks. Do you know of any good templates for standard work for Gemba walks?
Me: What are you trying to accomplish with Gemba Walks?
CI Leader: We know that Toyota leaders spend a lot of time at the Gemba. I have heard even very senior leaders spend at least 20% of their time at the Gemba. Our leaders are in meetings and fighting fires, but do not spend time coaching at the Gemba.
Me: So you want them to spend more time coaching?
CI Leader: At this point, I would be happy if they left their office and spent time at the Gemba.
Me: What do you want them to learn? What do you want them to teach?
CI Leader: You know, the kind of leader standard work we are looking for gives them guidance on what questions to ask, what to look for, and how to respond.
Me: Let’s go to the Gemba and see. Perhaps we can better understand what going to the Gemba might accomplish based on your current condition.
The point I’m trying to illustrate here is that there is no useful standard work templates that people can copy and paste. What’s needed is careful consideration of why you want leaders to do this and what you are trying to teach each of them right now. A Toyota sensei might ask: “What is the purpose?”
I think most would agree that getting leaders to walk the floor just so they are out there is not enough. Recall the “Management by Walking Around” movement popularized by In Search of Excellence. There are three main reasons for doing the Gemba Walk:
- Developing Leaders at the Gemba—One thing we learned from Toyota is the value of leaders who coach and develop others. We also learned that the most important work is at the Gemba so the most important coaching is of people doing their work or leading others at the Gemba.
- Leaders at the Gemba developing others—At the Toyota plant in the UK they have revamped their Floor Management Development System. This includes the visual board to reflect the KPIs and targets and state of the process and how it is used for daily meetings. These meetings are designated as a “window of opportunity” for coaching. For example, managers can observe group leaders in this controlled environment and give feedback.
- Overall management effectiveness—Leaders need to understand what is going on at the Gemba to effectively manage. “What is the current state of processes? What is the current state of leadership development? What do we need to work on next?”
Importance of the Routine – Leader Standard Work
Regardless of how creative we are and how variable our work demands are we all have repeating patterns in our day. We get our coffee, we check our emails, we go to meetings with standard agendas, we have habitual ways we make presentations, respond to questions from our subordinates, and so on. Leader standard work defines the best way we know today for the repeating parts of a leader’s day to be efficient and effective. They do not need to be recorded in detail with timing for each activity. In the Floor Management Development System at the UK plant, they define minimum role requirements for each leadership position in the hierarchy. This specifies a minimum to be achieved in a day. The portion of the day that is repetitive will be higher for those closest to the shop floor and lower for higher levels of management. For example, we can see in Figure 1 (What is leader standard work?) that the team leader in a Toyota plant, who is tied to their team, responding to andon calls, and making various types of checks, has up to 80% repetition, while the manager of a major area of the shop has 80% non-repeating activities. But even the manager has 20% of their day that can benefit from standard work.
What is leader standard work?
Leader Standard Work (LSW) is the repetitive pattern of activities that represent the current least wasteful method of planning and controlling normal business processes.
One type of leader standard work is for structuring Gemba Walks. Figure below shows a typical plan for a manager in a Toyota plant. They walk the floor looking for abnormalities. They want to confirm that the abnormalities are visible and being appropriately managed… and help if there is a need for resources beyond the capacity of the teams. In this plan, you see the leader has plotted out her walk pattern and noted what to look for at each stop. In certain areas, there will be a longer stop to do a deep dive. This is not generic standard work for all time, but rather what they are looking to confirm now based on the current condition in each area.
Typical Gemba Walk manager plan at Toyota
Leader standard work is greatly enhanced by good visual management. What visual control does is highlight the gap between actual and standard. At the worksite, this is the purpose of the andon which is a very active form of visual control. When the team member notices anything out of standard his job is to pull the cord. This causes a light to turn on and music to play and the team leader should arrive within 10 seconds. The team leader can contain the problem and override the line top or let the line stop. Then he records the problem. At the end of each day, the team leader and group leader review andon pulls and pick a problem to correct. Managers who come to the area can check that the andon system is being used properly and look at the types of problems the teams are having. Another great visual is the KPI board which reflects actual versus target. Standard work for each process is a great tool to audit and check how the work is being performed compared to the standard.
In short, leader standard work is a plan for the walk and what to look for and at each stop, the visuals are reviewed to compare plan to actual. This is one way to provide accountability for leaders in the area and instill discipline. A formal plan is most important at the beginning stage of developing the fundamental skills of walking the Gemba.
Leader Standard Work is a necessary step in the beginning stage until leaders develop so it is “the way they think and act”
How to Get the Most Out of Gemba Walks
Here is a quick summary allowing you to get the most out of your Gemba Walk:
- The purpose of Gemba walks is to develop leaders who develop their team members to continuously improve operations.
- Leader standard work can aid in making Gemba Walks effective and efficient.
- Unlike repetitive jobs in production, leader standard work is not doing the same thing every cycle, but a plan that must be modified to suit the current condition.
- The effectiveness of Gemba Walks is greater as processes are more stable and there is good visual management making abnormal conditions visible.
- Leader standard work is to get leaders started and, as they mature, it becomes a habit.
It should be clear by now that the structure for Gemba Walks is very situational. There is no ‘one size fits all’. It is different for beginners and those more seasoned, it is different depending on the lean maturity of the organization, and it is even different based on the current condition of each day.
This makes it difficult to plan and deploy. Many organizations seem to want to have one standard way so that all leaders will “get trained up” and then be effective at Gemba Walks. It simply won’t work this way.
I like the model Mike Rother came up with for planning activities like this, what Rother calls the improvement kata (see figure below). Rother says that when you want to accomplish something beyond your current knowledge threshold it is best to use the scientific method. Unfortunately, few of us are very good at it. We seem wired to make assumptions and to do it our way. In Rother’s model leaders learn through repetitive practicing of a standard pattern for working towards achieving major business challenges.
Using Kata to Develop Habits of Scientific Thinking
There is a role for a coach and a role for a learner. The learner is trying to reach a clearly defined challenge. Let’s consider the challenge of developing effective Gemba Walks. The challenge might be to make all managers from a certain level up effective at daily Gemba Walks within a year. Rather than search for a known solution, like a standard form for leader standard work, the person who is leading the planning effort (the learner) will follow the scientific process. A coach will meet with the learner regularly, ideally daily, to ask questions and identify any abnormalities in following the improvement kata pattern. What the coach wants the learner to do is:
- Understand the Challenge (the direction).
- Understand the Current Condition (current levels of standard work, of daily management, of visual management, of manager’s leadership skills, etc.).
- Establish the next Target Condition. This is a short-term target on the way to the challenge. For example, the first target condition might be to develop a standard work form for the Gemba Walk starting with one leader as a pilot.
- Experiment toward the Target Condition. The experiments test one factor at a time. They are rapid learning cycles. Plan your next step, test it, look at the results, and reflect on what was learned. This is PDCA at its best.
By the end of the year the chances of having a well—defined approach to training leaders to do Gemba Walks, while at the same time meeting the challenge of developing the managers, is very high. I would put my money on this approach over getting a standard template and trying to implement it.