A Tale of Two Sites

With apologies to Charles Dickens, but the opening line is just too good to resist…

The Best of Times

In this plant, the advance team is chaired and actively led by the most senior manager on the site. He is actively coaching, he is actively being coached. He is questioning his own learning, seeking council, and acting on it.

They are clear that, while there may be general guidelines, they must learn by trying and experimenting. They cannot simply deploy a roadmap because they can only see the next mile on a 1000 mile journey.

They see it as a method to shift their culture away from its “tell me what to do” legacy and toward one of an empowered workforce that takes initiative and works on the right things, the right way.

There is no doubt among the leadership team that this is the path forward.

They are starting to apply the language of the Improvement Kata informally in their meetings and discussions.

Overall, it seems a bit messy. But learning is like that.

The Other Site

The “implementation of Toyota Kata” is a directive from the corporate Continuous Improvement team.

The corporate team spends much of their energy developing and deploying templates, PowerPoint presentations, setting standards for the forms and the layout, lettering and colors on the improvement boards, and setting milestones.

They have published a step-by-step procedure for a site to implement Toyota Kata, based on their assumptions of what ought to work. None of them has actually led a change like this.

They are, in turn, working through the site continuous improvement team who is expected to execute to these standards.

The site leader receives weekly reports on progress. Training the managers and “implementing Toyota Kata” is the responsibility of one of the site’s continuous improvement staffers. The site leader questions him using the 5 Questions each week, and issues direction in response to the answers.

It is the continuous improvement practitioner who is responsible for motivating the members of the management team to challenge their own processes and develop their improvement boards. A significant number of them are questioning the need or purpose of this exercise.

Thoughts

Unfortunately I run into the second case far more often than I see the first. But the story is decades old. That is how we did Six Sigma, kaizen events, Theory of Constraints, Total Quality Management. In each case we have separated the deployment of a core change in the way we manage operations from the responsibility for actually managing.

It

doesn’t

work.

This TED talk by Tim Harford actually sums up the difference pretty well:

But beyond what works, and what doesn’t, we also have to ask “Which approach is respectful of people?”

What are the underlying assumptions about the people at the gemba when “standards” are established thousands of miles away, published, and then audited into place?

Why do they feel they must tell people exactly what to do?

What do they feel is lacking on the site?  Competence? or Clarity?

Notes From Day 2 of Kata-Con

Perfection is the enemy of progress.

  • The longer it takes, the higher the expectation.
  • The higher the expectation, the longer it takes.

My thoughts: I’ve seen this a lot. It is magnified when the leaders are detached from the process.

Process improvement is messy, and if the leaders aren’t comfortable with that messy process, they develop unrealistic expectations of what “progress” looks like.

The people getting the work done, meanwhile, end up working hard to manage those expectations. They actually conceal problems from the boss, for fear of him misinterpreting problems-that-must-be-solved with my-people-don’t-know-what-to-do.*

Trying to layer Toyota Kata over the wrong organizational structure will overwhelm people.

The organizational structure follows necessity. This lines up with Steven Spear’s research.

The organizational structure must match the needs of the process, and the target condition for learning.

If your supervisor has 20 direct reports, it is unlikely he will have the time to work on improvement in a productive way. Toyota’s team leader structure is specifically engineered for improvement, development, and getting a car off the line every 58 seconds.

 

Improvement takes time and people.

The End.

This isn’t free, nor can you calculate an ROI ahead of time. Get over it.

Start with what you MUST accomplish and look at what is required to get there. It doesn’t work the other way around.

If you don’t continually strive, you die.

If you aren’t striving to go forward, you are going backward.

My thoughts: I make the following analogy: Continuous improvement is like a freezer. There is never a time when you can say “OK, it’s cold enough, I can unplug it now.” You must keep striving to improve. Without the continuous addition of intellectual energy, entropy takes over, and you won’t like the equilibrium point.

All of our failures have come to good things.

My thoughts: By deliberately reflecting and deliberately asking “What did we learn?” you can extract value from any experience. The way I put it is “You have already paid the tuition. You might as well get the education.”

We had sponsorship challenges as the leaders caught up with the people.

My thoughts: Yet another instance of the leaders falling behind the capability of their people. When the people become clear about what must be done, and just start doing it, the only thing an uninformed leader can do is either get out of the way or destructively interfere.

People don’t like uncertainty. Kata deliberately creates uncertainty to drive learning. You have to be OK with that.

My thoughts: Another expression of the same point from yesterday.

“Learning only” has a short shelf life.

“Cool and Interesting” is not equal to Relevant.

Those are the words I wrote down, rather than the words I heard. The key point is that you can, for a very short time, select processes to improve based on the learning opportunities alone. But this is extra work for people. The sooner you can make the results important the quicker people get on board.

A business crisis should not stop improvement or coaching. Does it?

My thoughts: This is a good acid test of how well you have embedded. When a crisis comes up, do people use PDCA to solve the problem, or do they drop “this improvement stuff” because they “don’t have time for it.” ?

Inexperienced 2nd coaches coaching inexperienced coaches coaching inexperienced learners… doesn’t work.

A lot of companies try to do this in the interest of going faster. Don’t outrun your headlights. You can only go as fast as you can. Get help from someone experienced.

Just because you have gone a long way doesn’t mean you can’t slip back. You must continue to strive.

The “unplug the freezer” analogy applies here as well.

You don’t have to start doing this. But if you choose to start, you may not stop. You have to do it every day.

Don’t take this on as a casual commitment, and don’t think you can delegate getting your people “fixed.” (they aren’t broken)

Everybody gets it at the same level. Senior managers tend to lose it faster because there is no commitment to practice it every day at their level.

Awareness is a starting point, but not good enough. A 4 hour orientation, however, is not enough to make you an expert… any more than you can skim “Calculus and Analytic Geometry” and learn the subject.

Results do get attention.

“I’ll have what she’s having”

But don’t confuse results with method.

My challenges to the plant managers weren’t about P&L or service levels. They were about moving closer to 1:1 flow, immediate delivery, on demand.

Challenges must be in operational terms, not financial terms.

Move from “These are the measures, and oh by the way, here is the operational pattern” – to –> “This is the operational pattern I am striving for. and I predict it will deliver the performance we need.”

Gotta catch a plane. More later.

 

______________

*When I was in the Army, we got a new Battalion Commander who listened to the logistics radio net, where the staff officers discussed all of the issues and problems that had to be solved. He would jump to a conclusion, and issue orders that, if carried out, would interfere with getting those problems solved.

Although he spoke of initiative and taking action, his actions revealed he wasn’t willing to trust us to let him know if there was a problem we couldn’t handle, and expected perfection in execution in situations that were chaotic and ambiguous.

We ended up finding an unused frequency, and encrypting our traffic with a key that only we shared, so the commander couldn’t hear us. Yup… we were using crypto gear, designed to keep the Soviets from hearing us, to keep our boss from hearing us.

As the information channels to him slowly choked off, he was less and less informed about what was actually happening, and his orders became more and more counter-productive, which in turn drove people to hide even more from him.

This, I think, is a working example of “getting bucked off the horse.”

Notes From Day 1 of Kata-Con

I’m attending the Toyota Kata Summit in Fort Lauderdale. We’ve completed Day 1. Here are some notes I took. I plan to expand on some of them later.

Experiment your way forward vs. decide your way forward.

My thoughts: We like certainty. We like a plan we know will work. Unfortunately a plan we know will work is usually just a plan we have convinced ourselves should work. But we have a hard time distinguishing the difference. Such is our desire for certainty – a strongly held opinion or feeling becomes a fact.

We don’t know how it will go. We have to get comfortable with ambiguity if we want to move beyond what we already know.

Until you understand the mindset of uncertainty, you can’t teach anyone else.

My thoughts: This is probably the most foundational qualification for a manager who aspires to become an improvement coach. The improver / learner won’t know the answers.. and neither will you. You have to be very OK with that. If you aren’t, then you’re just telling people what to do, and nobody is learning anything.

Until you try, there is no baseline for learning or coaching.

My thoughts: At the start of a ski lesson, the instructor had each of us ski 50 yards or so down the hill while she watched from below. She asked us to just ski down to her. Of course, she observed.

Then each of us was individually coached on something to practice… a drill or technique that would correct one deficiency in our form.

She skied down another 50 yards, and observed us again. If we were not performing the drill correctly, she corrected until we were performing it correctly. Then our task was to practice.

Asking “What is your target condition?” is asking the learner to demonstrate his technique and understanding. For the coach, the idea is to get them to try so you have a baseline for coaching.

Warning! Be ready for empowered employees!

If a manager with only an “awareness” level of understanding enters this space, he’ll either deflate the team with inept coaching, or will get bucked off the horse he isn’t good enough to ride. The leaders can’t do this from behind. You can’t simultaneously have empowered, creative team members and maintain control. (Sounds a bit like something Heisenberg came up with…)

I am coaching (by giving direction).

No you aren’t.

My thoughts: Actually this note was my thought triggered by something someone said. But I’ve seen senior managers who want to be “coaches” and then redefine “coaching” to mean “tell people what to do.” Doesn’t work like that.

The 5 questions are a jig. The questions on the card allow you (the coach) to listen because you don’t have to think about what question to ask next.

Some days big up. Every day little up. Please try. Do your best. Until you take first step, you cannot see next step. – Toyota coordinator (master coach).

How Does Kaizen Differ From A Kaizen Event?

The title of this post is a search term that hit the site today. It’s an interesting question – and interesting that it gets asked.

“Kaizen” is now an English word (it’s in the OED) and defined as such:

Definition of kaizen in English:

NOUN

A Japanese business philosophy of continuous improvement of working practices, personal efficiency, etc.

Origin

Japanese, literally ‘improvement’.

 

Let’s talk a bit about that “Japanese, literally ‘improvement” bit.

Jon Miller of the Kaizen Institute was raised in Japan and offers up this nice breakdown of the meaning behind the meaning:

image

Jon explains:

•Kai = Change is made from two characters “self” and the picture “to whip”. You can see the stripes across the poor fellow’s back.So Change is something you do to yourself.

•Zen = Good. In this case a sacrificial lamb which means “righteous” is offered between two characters for “word”. In this case word = clear and precise speech. So good sacrifice with precise speech all around it is “Good”. The Zen character we are using today is a simplified version. Also, Zen in this case is not the same character as Zen Buddhism.

•Kaizen = Change for the better. Or in our case whip yourself so that you can make a nice sacrifice and always have clear speech / thoughts surrounding you. Smile

As I understand it, in vernacular Japanese, “kaizen” is regarded as a business term. The word is not used in day-to-day context.

From Jon’s explanation, there is also a very personal aspect to it. Change is something you impose on yourself.

With all of that, “kaizen” is simply a word that generally refers to any systematic disciplined activity of improvement.

Kaizen Events

Now things get tricky, because here in the West, we have often regarded “kaizen events” and “kaizen” as the same thing. They aren’t. While you can certainly make improvements with kaizen events, that isn’t the only way to improve things, and I’ll add it isn’t necessarily even the best way.

A typical Western kaizen event (and there are lots of variations, though most companies that use them tend to impose fairly rigorous attempts to standardize them) is a 5 day focused effort with a 100% dedicated team.

The kaizen event leader is usually a specialist whose job is to plan and lead these things, identifies an improvement opportunity. He might be tasked by shop floor management to tackle a chronic or painful problem, or might be executing the “lean plan” that calls for a series of implementation events.

It is his job to plan and execute the event and to bring the expertise of “how to make improvements” to the work force and their leaders.

Here’s the Problem

The full-time kaizen event leaders typically get really good at seeing improvement opportunities, organizing groups for improvement, and quickly getting things done. They get good at it because they do it all of the time.

The area supervisors might be involved in a kaizen event in their area a few times a year if that. Some companies target having each employee in one kaizen event a year.

That’s 40 hours of improvement. All at once. The question is: What do they do (and learn) the other 1900 hours that year?

What do they do when something unexpected happens that disrupts the flow of work? Usually kaizen events don’t deal with how to manage on a day-to-day basis other than leaving an expectation for “standard work” in their wake.

But “standard work” is how you want the work to go when there aren’t any problems. When (not if) there are problems, what’s supposed to happen?

This is why many shop floor leaders think “kaizen” is disconnected from reality. Reality is that parts are late, machines break, things don’t fit, Sally calls in sick, and the assembler has to tap out threads now and then. In the hospital, the meds are late, supply drawers have run out, and there is a safari mounted to find linens.

These things are in the way of running to the standard work. They are obstacles that weren’t discovered (or were glossed over as “resistance to change”) during the workshop.

The supervisor has to get the job done, has to get the stuff out the door, has to make sure the patients’ rooms are turned over, whatever the work is. And nobody is carving out time, or providing technical and organizational support (coaching) to build his skills at using these problems as opportunities for developing his improvement skills, and smoothing out the work.

So… sometimes, “kaizen events” implement changes, but don’t necessarily do much for that personal drive for improvement.

The Shift

We’re starting to see a shift, as we realize those front line leaders are actually the ones who need to be the experts. That’s where Toyota Kata comes in – as a tool to learn to coach those leaders so they learn to keep the improvement going rather than just fighting erosion and working around problems.

OK – it’s late, and this was actually a diversion from something else I’m working.

Goal vs “Target Condition”

Emily sent an email asking “how would you describe the difference between GOAL and TARGET CONDITION?”

I end up on this topic enough that I thought I’d discuss it here.

I am assuming we are referring to the “Toyota / Toyota Kata” context here. I mention that because while “target condition” has a pretty clear meaning in that context, we have to rely on the everyday meaning definition for “goal.”

Thus, I can’t objectively say “goal” means this, or doesn’t mean that because it means whatever it means to you.

Still, I can give it a try.

I have drawn on the following analogy frequently because I think it demonstrates the concept pretty well.

“I believe that this nation should commit itself to achieving the goal, before this decade is out, of landing a man on the Moon and returning him safely to the Earth.”

I’d say that was a goal. It is a specific accomplishment that is clearly “done” or “not done” and it has a deadline. (Someone else once said “a goal is a dream with a deadline.”)

For those of you who don’t remember Woodstock*, let me provide some context.

President Kennedy made that speech on May 25, 1961.

imageOn May 5th, Alan Shepard had been launched into a sub-orbital space flight giving the USA a total manned space flight experience base of about 15 minutes. The previous month, the Soviet Union had launched Yuri Gagarin for a single orbit around the Earth, thus the entire planet had a total manned space flight experience of just under 2 hours (but the Russians weren’t sharing).

<— This is the best we could do.

The goal was selected for political and technical reasons. We had developed a huge rocket engine needed to do the job, and didn’t think the Russians larger rockets would scale well. So the Administration selected a goal they thought the U.S. could accomplish but believed the USSR would have a tougher time with. (They were right.)

At the time the speech was made, there were two competing approaches in play for landing a man on the moon.image

Both involved landing a large upper stage intact on the moon, then lifting off and using the whole thing to return to Earth.

The problem was thought to be how to get that huge moon landing rocket off the Earth and to the moon.

There were a couple of ideas kicking around at the time.

One was to build a huge rocket, maybe twice the size of the Saturn-V that eventually was used for Apollo 11.

imageThe design was never finalized, but the concepts were all lumped together as the “Nova” rocket. You can see the 36 story tall Saturn V (the biggest rocket ever built and launched) as the second from the right in this picture. The idea was to send the entire thing directly to the Moon with a single launch. This was called the “Direct” approach.

Given that building the Nova rocket (not to mention the launch facility) was likely to be…um…really hard, the other idea was to use multiple launches of something more like the Saturn rocket, and assemble the moon rocket in low Earth orbit, then send it on its way. This was called “Earth Orbit Rendezvous.

All through the late spring and summer of 1961 this debate was raging within NASA. Wernher von Braun, our chief “rocket guy” wanted this capability for a large lunar payload because he was interested in establishing bases and serious exploration of the moon. But that wasn’t the objective right now. It was get there fast and beat the Russians.

Another NASA engineer, John Houbolt had what was considered a bit of a high-risk (bordering on crackpot) scheme of a smaller-but-still-huge rocket, single launch, sending an expendable two-stage lander to the moon, having it land with two astronauts, then lift off and rendezvous with the return ship in lunar orbit. Not surprisingly this scheme was known as “Lunar Orbit Rendezvous.” It was risky because what was thought to be the trickiest part, the rendezvous and docking, was to be done 250,000 miles from the safety of Earth, with no way home if it didn’t work.

You can read the whole story here.

By the fall of 1962, Lunar Orbit Rendezvous had emerged as the only viable scheme to accomplish the goal by the deadline.

At the program level, they now had a target condition: How the process should operate in order to accomplish the goal. This does not mean they had worked out every detail. It only means they knew what they were trying to accomplish. This 1962 NASA film pretty much lays out the concept in as much detail as they understood at the time:

Remember, this is at the high level. But at this level they identified obstacles – things they didn’t know how to do – that had to be cleared in order to reach the target condition.

1) They had to develop the concept into a real rocket capable of pushing 90,000 pounds (and finally 100,000 pounds) into lunar orbit. They also had to develop and built the infrastructure to (according to the initial plan) launch one of these every month.

2) They had to determine if (and how) humans could spend 10+ days in space without psychological or psysiological problems. (Remember, we had no idea at the time).

3) They had to develop a space suit that would allow an astronaut to leave the spacecraft.

4) They had to develop techniques and technology for rendezvous and docking in orbit.

Each of these major obstacles could be, in turn, defined as a goal (or challenge) for the next level down. Project Gemini’s purpose was to test #2, and directly learn about #1 and #2.

imageimage

 

 

 

 

 

 

 

 

And of course, the teams working on developing space suits, developing docking technology, etc. then would set their own target conditions that progressively marched toward their goals or deliverables.

Bring this back to Earth

A goal is something you need to accomplish. It usually doesn’t assign the method, only the result and the “by when.”

A target condition is typically a major intermediate step toward the ultimate challenge. Importantly, it outlines the “how” or proposed approach, though necessarily doesn’t offer up the solutions to the problems.

The goal is “win the game.” The target condition is the game plan. To execute the game plan, we need to develop specific capabilities, or solve specific problems.

One thing the target condition does do is limit the domain of the problems that must be solved. This is critical.

There are always more problems to solve than are solvable with the resources available. By being specific about a target condition, you focus the effort on the obstacles that are actually in the way of achieving the target. As you proceed, you learn, and as you learn, the nature of those obstacles may change. Thus, the obstacles are not a static list of things to do. Rather, they are the unsolved issues that, right now, you think must be dealt with.

See this (largely redundant) post from a couple of years ago for another perspective. (oops – just realized I’d already used the Apollo analogy. Oh well. At the time all of this was going on, I was the geeky 12 year old who knew (and would talk endlessly about) every dimension, rocket engine nomenclature, fuel burn rates, etc. of the Saturn-V rocket.)

Hopefully this will spark some discussion.

———–

*If you actually remember being at Woodstock, then you likely weren’t there.  Winking smile

Shifting Perspective from Getting Results to Developing People

Note: This post has been in my “Draft” queue for a few months, so actually pre-dates the previous one. But I’m seeing a theme developing in what I am paying attention to lately.

Taken from an actual conversation.

“What is your target condition?”

“To get [this productivity metric] from 60% to 85%.”

(thinking) – he is only talking about the performance metric.

“How would the process have to work to achieve that level of performance?”

(pointing to process flow diagram) “We have two work flows. One for routine project work, the other is high-priority emergent work. Whenever a worker has an opportunity to take on routine project work, I want him to be able to take the next most important job from the prioritized work queue.”

“This would eliminate the need for the worker waste time trying to find me to get an assignment or investigate or guess what he should do next, and let him get started right away. It would also stop cherry picking the easier jobs”

“My goal is for the Team Leader to establish those priorities, and keep track of how work is progressing.”

(thinking) – OK, I see where he is trying to go. I’m not sure I would have taken this exact approach, but it looks good enough right now to let him run with it and see what he learns.

“What was the last step or experiment you completed?” (Note: I am trying asking this question with ‘you completed’ so emphasize I want results from the last one, not information about what is ongoing or next.)

“Before I went on vacation, I looked at the incoming work queue, and established priorities for that work. As the workers needed their next job, it was clear to them what they needed to do.”

“What result did you expect?”

“I expected my productivity metric to hit my 85% target.”

“What actually happened?”

“The metric did hit the 85% target, when I prioritized the work. Then I went on vacation, and as you can see here (pointing at the graph), the productivity dropped back to its baseline level.”

“What did you learn?”

“I learned that when I set the priorities, and track the productivity, it improves as I expected it to. I also learned that when I don’t do it myself, then things go back to the way they were.”

“What obstacles do you think are preventing you from achieving your target condition?”

“People don’t know what the most important job is.”

“You said your target condition is for your team leader to make those assignments. How does that obstacle relate back to your target?”

“My team leader doesn’t know what the most important jobs are.”

“How about writing that down on the obstacle list.”

(he adds it to the list)

“Any other obstacles?”

“Um… I don’t think so.”

(thinking) I’m pretty sure there are other issues, but he seems focused on this one. Let’s see where he is going with it.

“OK, so that (the team leader) is the obstacle you’re addressing now?”

“Yes.”

“OK, what is your next step?”

“I am going to assign the priorities myself.”

“What result do you expect?”

“I expect the performance to go back to its target value.”

“How is that addressing the fact that your team leader doesn’t know how to assign priorities?”

pause…

(thinking) He isn’t connecting the dots here.

“Let’s come back to your target condition. You said you want the team leader to do all of this, rather than you, right?”

“Yes.”

“What is keeping you from just giving him the work and having him do it?”

“If I did that the productivity would go down again.”

”How come?” (Yes, I am leading the witness here. My thinking is that if I can get the right words to come out of him, he’ll “get it.”)

“My team leader doesn’t know how to set the priorities.”

“Right, so that’s the real obstacle in the way of reaching your target of having him do it, right?”

“Yes, but if I have him do it, then my productivity will go down. Isn’t the idea to hit the goal?”

“Yes, but you said in your target condition you wanted to hit the goal by having your team leader set the priorities, not just do it yourself.”

pause… (he’s probably feeling a little trapped now.)

(continuing) “So, if your team leader did know how to set the priorities, you think you’d hit your productivity goal with him doing it, right?”

pause…

“What else might be in the way?”

“He (the team leader) doesn’t like to go and talk to managers in other [customer]departments about their work priorities.”

“Write that down. Anything else?”

“He sometimes is reluctant to give assignments to people who would rather be working on something easier [but less important].”

“Write that down. Anything else?”

“I don’t think so. It sounds like the team leader is the problem.”

“You remember the session we did about David Marquet, the submarine captain, right?”

pause… “yeah.”

“So is this an issue with his (the team leader’s) competence – something you need to teach – or clarity – something you need to communicate?”

“I guess I need to teach him how to set the priorities.”

“So that is still the obstacle you are addressing now?”

“Yes.”

“OK, so what step are you going to take first?”

and from here, the conversation took a 90 degree turn into how this manager was going to develop his team leader.

The target condition got clarified into the capabilities and information the Team Leader needed to be able to perform the job competently.

The obstacles turned into things which must be taught, and things which must be communicated.

In retrospect, the obstacle I was addressing was reluctance on the Manager’s part to accept that developing the team leader was nobody’s job but his. But I’m finding that to be a common theme in a few places.

Does Your Solution Have A Problem? Does Your Problem Have A Customer?

Javelin.com is a site with a few good tools centered around startup product development. (“Lean Startup”). I really liked their tutorial around the “javelin board” which is a vertical PDCA record specialized for testing product ideas.

In the tutorial, the phrase that really got my attention was this:

“Not all solutions have problems, and not all problems have customers.”

If you are a regular reader, you know one of the questions I ask frequently is “What problem are you trying to solve?” This is especially important if the proposed solution is a “lean tool.” For example, “there is no standard work” is not a problem, per se. I know lots of companies that do just fine, and have more than doubled their productivity before work cycles ever emerged as something to work on. “What obstacle are you addressing now?” is a question we ask in the Coaching Kata to explore the learner’s linkages between the proposed solution, the problem (obstacle), and target condition. The obstacle itself is a hypothesis.

The javelin board process first ensures that (1) you know who the customer is and (2) that you validate that the problem you THINK they have is one they ACTUALLY have… before you go exploring solutions.

Remember as you watch this, though, that the process isn’t different from the Improvement Kata. It is just a specialized variant. The underlying thinking pattern is totally identical… and the problem Toyota Kata is trying to solve is “We have to learn this thinking pattern.” Once you understand the pattern, and apply it habitually, then these variations make perfect sense. On the other hand, if you don’t understand the underlying pattern, then these variations all look like a different approach, and you’ll end up wrestling with “which one to adopt.”

DMAIC and Toyota Kata

A lot of the organizations I deal with have a legacy with Six Sigma, or some x-Sigma variant. If they are now trying to incorporate Toyota Kata as a way to shift their daily behavior, questions arise about how it fits (or might fit, or whether it fits) with DMAIC.

This sometimes comes about when the impetus to embrace Toyota Kata comes from outside the organization, such as an initiative from the corporate Continuous Improvement office. In this case, unless integration with legacy approaches is carefully thought through, Toyota Kata (or whatever else is coming down the pipeline) can easily be perceived as “yet another corporate initiative” or “something else to do” rather than “a new (and hopefully better) way to do what we are already doing.”

My Background

First a disclaimer. My deepest exposure to Six Sigma was during my time as a Quality Director in a large company that had a long history with TQM and then Six Sigma. Thus, I dealt with the Black Belts and Green Belts in the organization, and paid a lot of attention to the projects they were working on.

In addition, every certification project for a new Black Belt came across my desk. Unlike a lot of managers in the chain (apparently), I actually read them, parsed them, and asked questions when I couldn’t follow the story line of the project. (Apparently nobody expected that, but it’s another story.)

I worked with the corporate Master Black Belt, made input into their programs, and did what I could to create a degree of cooperation, if not harmony, between the Six Sigma community and the lean guys.

Thus I am not claiming this is anything new or profound. Rather, this is sharing my own sense of connection between these two approaches in a world where I often find them competing for people’s mindspace.

The Improvement Process Flow

As I observed it, a Six Sigma project was typically organized and conducted as follows:

An area manager, usually a Green Belt, identifies something that needs improving. He assembles a team of stakeholders. He is coached by a Black Belt through*:

Defining the problem and establishing a charter for the team.

Establishing a Measurement that will define progress.

Conducting a thorough Analysis of the process, with a primary focus on sources of variation, especially those which are intertwined with quality issues.

Developing a list or set of Improvements and putting them into place, again focusing primarily on variation in methods, etc. that drive defects.

Establishing a standard to Control the process and keeping it running the new way.

Define, Measure, Analyze, Improve, Control. DMAIC.

In actual practice, it is very similar to the large single-loop “six step problem solving” process I was taught as “the problem solving process that came out of TQM.” That is probably not a coincidence.

DMAIC projects are typically targeted at measurable significant financial payback. Black Belts are trained to find and spend their time on high-payback projects. At least in the company I worked for, there was a minimum payback they had to achieve to get their certification. They also had to demonstrate knowledge of the various high-powered statistical tools.

It makes sense for people steeped in this methodology to ask how it fits in with “short cycles of experiments and improvements” that are the anchor for not only Toyota Kata, but kaizen in general (if you are doing it right).

And, put another way, if the organization is trying to establish a coaching culture using Toyota Kata, is Toyota Kata something different from DMAIC, or do they fit together? (I have actually been asked exactly the same question about Toyota Kata and kaizen(!), but that, too, is another story.)

To give credit where credit is due, this was the topic over lunch last week at a client site whose Six Sigma projects also follow the general structure I outlined above. Jazmin, the continuous improvement leader, was already working through this in her mind, and recognized the linkage right away.

Since the company has an active Six Sigma program, with dozens of projects ongoing, we wanted to find a way to integrate Toyota Kata thinking into what they were already doing vs. introducing yet another separate initiative. (It is easier to “embrace and extend” something you are already doing than bring something brand new into the domain.)

Relating DMAIC to the Improvement Kata

Here is how I relate DMAIC to the Improvement Kata.

Define the Problem =(more or less) Challenge and Direction. This is what we are working on, and why it is important.

Measure and Analyze = Grasp the Current Condition. Six Sigma has a host of powerful tools (which are often used just because they are there … so be careful not to make easy things complicated).

I would point out that if you follow the process in the Improvement Kata Handbook, you are also initially focused on variation in the process. Lean people tend to reduce all variation down to units of time, but in that noise are all sources of variation of the process. Defects, for example, don’t count as a delivery, and so introduce noise into the exit cycles. Machine slowdowns and stoppages, likewise, disturb the rhythm of the process. Like DMIAC, the Improvement Kata, as outlined in the handbook, steers you toward sources of variation very quickly.

Thus, a Six Sigma project team, and an improver following the Improvement Kata are both going to initially look for sources of instability. (Quality First, Safety Always).

At this point, the two diverge a little, but only a little.

Perhaps because DMAIC sounds like a single cycle, a fair number of teams tend to try to Implement a Single Grand Solution. They spend a fair amount of time brainstorming what it should look like, and designing it. Then, once they think they have a solution, they put it into place by establishing new “standards” (in this context, that usually means procedures), training people, and validating that it all works.

Again – a lot of kaizen events do pretty much the same thing, they just might do it faster if it is a classic five-day event.

A few years ago I was on a discussion panel at a conference in Chicago that was very Six Sigma centric. In the various breakout sessions, the Black Belts (who are mostly staff practitioners) universally complained about “management embracing the changes” and not enforcing the new processes. They were frustrated that once they were done, things slipped back.

In other words, once the energy input of the project itself stopped, entropy took over, and things regressed back to the original equilibrium.

And (once again) traditional kaizen events often have the same problem.

Blending Toyota Kata and Six-Sigma Coaching

Think of this conversation between the Black Belt coach and the Green Belt project leader in their daily meeting and check-in:

[preliminary social rituals]

“Just to review, what is the problem you are working on?”

[Green Belt reviews the charter and objective]

“Great, so what is the target condition you are striving for right now?”

[Green belt describes the next intermediate step toward the chartered defined state. He describes how the process will operate when key parts of it are stabilized, for example.]

The Black Belt is listening carefully, and may ask follow-up questions to make sure the target condition is clearly on the path toward solving the defined problem vs. chasing something interesting-but-irrelevant to the issue at hand.

“Good. Can you tell me the last step you took?”

[Green Belt describes a change they made, or some additional data they needed to collect and analyze, or a control measure they have experimented with, etc]

“What did you expect from that step?”

[Green Belt reviews the intent of the action, and what he expected to happen.]

“And what actually happened?”

[Green Belt describes what his data collection and observation revealed about the process, or how well his control measure worked to contain a source of variation, etc…. or didn’t.]

“And what did you learn?”

[Green Belt describes insights that have been gained, especially insights into sources of variation, or the effect of controlling them. He also describes what he might have learned about the tools, or struggled with in applying them.]

“Very good. So what sources of variation or obstacles do you think are preventing you from reaching the target right now?”

[Green Belt describes the current suspects for causes and sources of variation in the process, as well as other issues that may be impeding progress.]

“Which one of those are you addressing now?”

[The Green Belt may well still be working on the last one, or might have it effectively controlled and is now addressing a new one. If he has moved on to a new one, the Black Belt is going to be especially interested in the control mechanism for the sources of variation that has been “eliminated” so it stays that way.]

For example, a follow-up question might be “Can you tell me how you are controlling that? What countermeasures do you have to detect if that variation comes back into play?”

We aren’t limited to SPC of course, and actually would rather have a binary yes/no  need-to-act/don’t-need-to-act signal of some kind.

What we are going here is iterating through sources of variation, and establishing a positive Control on each of them before moving on rather than trying to stabilize everything in one step at the end.

Once he is satisfied that the project team hasn’t “left fire behind them,” then the Black Belt can move on.

“Great. Can you tell me the next step you plan to take, or experiment you’re going to run?”

[Green Belt reviews the next action to either learn more about a source of variation or attempt to keep it controlled.]

“And what result do you expect?”

[If the Green Belt is proposing using a specific Six Sigma statistical cool, the Black Belt is going to be carefully listening, and asking follow-on questions to confirm that the Green Belt understands the tool’s function and limitations, how he plans to use it, what he expects to learn as a result, and why he thinks that specific tool will give him the answers he is looking for.]

“OK, when do you think we can review what you have learned from taking that step”

This interaction is using the Coaching Kata script to develop the Six Sigma skills of the Green Belt project leader. We already have a coaching relationship, so all we are doing here is practicing a technique to make it more effective and more structured.

DMAIC is now more like:

DM [repeat AIC as necessary]

as the team works methodically through the sources of variation as they are uncovered.

The Master Black Belt

At the next level up is typically a Master Black Belt who is generally responsible for mentoring and developing the Black Belts. They typically meet weekly or monthly and review and share progress on projects.

Only now, let’s shift the discussion to reviewing and sharing progress on developing the problem analysis and solving skills of the Green Belts. Remember, the Green Belts are line management, and we want to get them thinking this way about everything.

“Lets review your learning objective for your Green Belt last week.”

“What did he do? What did you expect him to learn?”

“What did he actually learn? Did he make any mistakes?”

“Is he stuck anywhere? What is your plan to give him additional coaching or instruction?”

In other words, they are following DMAIC as well. Except that the “problem” is the skill of the Green Belt and how effectively he is applying that skill to solve his chartered problem.

 

I am really interested in hearing from Six Sigma folks out there about how this resonates, or doesn’t, with you.

————

*Sometimes I observed a Black Belt doing the same thing on his own initiative, leading the project himself. Occasionally I saw a Black Belt acting solo. I am not discussing either of those approaches here, however.

Toyota Kata: Obstacles Are Not Action Items

Continuing my observations about old patterns that get in people’s way as they try to practice Toyota Kata…

“What obstacles do you think are preventing you from reaching the target?” (I also like to insert the word “now” in there sometimes to emphasize what I am writing about here.)

The intent (and to be clear, as I am interpreting it) here is for the improver / learner to give some thought to the things that have to change before the target condition can be reached.

It is not a list of stuff to do.

I am speculating here, but I suspect that there is a legacy from (mis)use of “Kaizen Newspapers” since, in actual practice, they often look similar to obstacle lists. (I believe the original intent of the “Kaizen Newspaper” was to function more like the Toyota Kata PDCA record, but that’s a topic for another post.)

Remember that there are two routines defined in “Toyota Kata,” the Improvement Kata and the Coaching Kata.

The Improvement Kata has four major steps, and the “five questions” that get the popular attention are actually part of the Coaching Kata.

Yes, it is the Coaching Kata that managers actually have to learn, but to be any good at it they (hopefully you!) need to fully understand the improvement kata so they can effectively coach a poor answer to one of the questions into a good answer. Without that baseline skill, the “5 questions” exercise turns into asking the question, waiting for the other person’s lips to stop moving, ask the next question. Doesn’t work (which is why I didn’t say “learner” – because no one is learning in this scenario).

Once the target condition is established, the improver / learner should then set back to the perspective of the current condition and assess what is keeping her from just moving directly to the target.

With all of that as an introduction, let’s look at Obstacles in context.

This is Mike Rother’s schematic of the Improvement Kata from the Improvement Kata Handbook (<– click on the link to download it from his web site).

image

In this model, “obstacles” are the things in the way of getting to the target condition. Saying the same thing, with more emphasis, “obstacles” are only the things in the way of getting to the next target condition.

But obstacles are not action items.

Obstacles are the issues, problems, etc. that you (as the improver) see at the moment. They are based on your current understanding of the current condition.

As you progress by running your experiments, or gaining more information, you are changing your understanding of the current condition.

That is why we ask “What is the actual condition now?” rather than “What was the condition you started from.

The current condition is just that. (Which is why, as a coach, you should insist that your improver / learner keep the current condition up to date.)

As your understanding changes, your view of what might, or might not, be an obstacle changes.

Obstacles are sometimes rendered irrelevant. You might have eliminated a troublesome process step entirely. Or solving one problem might have done “collateral damage” and taken out a couple of others.

As you advance your knowledge, you will likely change your understanding of the obstacles, reword them, clarify them, edit them. New ones might come into view (for the moment).

“What did you learn?” is often at least partially answered by pointing out changes in the obstacle list.

Obstacles are not action items because it is unlikely you will have to deal with all of them.

So… don’t worry about getting the obstacle list “right.” It’s just a reflection of your current understanding which can, and should change as you go.

Jeff Liker: Is Lean a Waste Elimination Program or Striving for Excellence?

Jeff Liker asks (and answers) the title question in a great Industry Week blog article by the same title.

One of the biggest obstacles we in the lean community need to overcome is our own inertia around “Lean is a process for finding and eliminating waste.”

In the article, Liker brings up a point that is often lost on us: Looking for the problems and negative things kills morale.

The “waste” that you see is the result of underlying issues and culture. Stop overproduction in one place, and it either returns or pops up elsewhere because the underlying reasons for it were never addressed.

Operations that are not operating at a high level of lean typically are lacking underlying process discipline, which leads to these problems and they proliferate daily as the company is in a constant firefighting mode. Trying to eliminate waste in the current system and culture is like identifying and fighting problems—it is debilitating and a losing proposition.

[Emphasis added]

I bolded that phrase for a reason.

We aren’t talking about a technical implementation here. We are talking about a shift in the underlying culture – the habitual ways people interact with one another, with the process, respond to challenges and problems.

Today we have, thanks to Jeff Liker and a few others, an excellent picture of an ideal version of Toyota. We know what it looks like.

Getting there is an entirely different proposition, as most companies that have tried this stuff know first hand. It is hard.

What is beginning to emerge, though, is that the thinking pattern that is learned through solving these problems the right way (vs. just implementing tool sets) is the same thinking pattern required to shift the culture.

It is hard. You have to do the work. But the way to get there is emerging.