Posts

Posted on

Prediction Doesn’t Equal Understanding

Lunar Eclipse over Everett, WA. Photo by Mark Rosenthal, © 2015Sometimes people fall into a trap of believing they understand a process if they can successfully predict it’s outcome. We see this in meetings. A problem or performance gap will be discussed, and an action item will be assigned to implement a solution.
Tonight those of us in the western USA saw the moon rise in partial eclipse.

We knew this would happen because our understanding of orbital mechanics allows us to predict these events… right?

Well, sort of. Except we have been predicting astronomical events like this for thousands of years, long before Newton, or even Copernicus.

The photo below is of a sophisticated computer that predicted lunar eclipses, solar eclipses, and other astronomical events in 1600BC (and earlier). Click through the photo for an explanation of how Stonehenge works:

Photo of Stonehenge
Creative Commons flickr user garethwiscombe

Stonehenge represented a powerful descriptive theory. That is, a sufficient level of understanding to describe the phenomena the builders were observing. But they didn’t know why those phenomena occurred.

Let’s go to our understanding of processes.

The ability to predict the level of quality fallout does not indicate understanding of why it occurs. All it tells you is that you have made enough observations that you can conclude the process is stable, and will likely keep operating that way unless something materially changes. That is all statistical process control tells you.

Likewise, the ability to predict how long something takes does not indicate understanding of why. Obviously I could continue on this theme.

A lot of management processes, though, are quite content with the ability to predict. We create workforce plans based on past experience, without ever challenging the baseline. We create financial models and develop “required” levels of inventory based on past experience. And all of these models are useful for their intended purpose: Creating estimates of the future based on the past.

But they are inadequate for improvement or problem solving.

Let’s say your car has traditionally gotten 26 miles-per-gallon of fuel. That’s not bad. (For my non-US readers, that’s about 9 liters / 100 km.) You can use that information to predict how far a tank of fuel will get you, even if you have no idea how the car works.

If your tank holds 15 gallons of fuel, you’ll be looking to fill after driving about 300 miles.

But what if you need to get 30 miles-per-gallon?

Or what if all of a sudden you are only getting 20 miles-per-gallon?

If you are measuring, you will know the gap you need to close. In one case you will need to improve the operation of the vehicle in some way. In the other case, you will need to determine what has changed and restore the operation to the prior conditions.

In both of those cases, if you don’t know how the car operated to deliver 26 miles-per-gallon, it is going to be pretty tough. (It is a lot harder to figure out how something is supposed to work if it is broken before you start troubleshooting it.)

Here’s an even more frustrating scenario: On the last tank of fuel, you measured 30 miles per gallon, but have no idea why things improved! This kind of thing actually happens all of the time. We have a record month or quarter, it is clearly beyond random fluctuation, but we don’t know what happened.

The Message for Management:

If you are managing to KPIs only, and can’t explain the process mechanics behind the measurements you are getting, you are operating in the same neolithic process used by the builders of Stonehenge. No matter how thoroughly they understood what would happen, they did not understand why.

If your shipments are late, if your design process takes too long, if your quality or customer service is marginal, if the product doesn’t meet customer’s expectations, and you can’t explain the mechanisms that are causing these things (or the mechanisms of a process that operates reliably and acceptably) then you aren’t managing, you are simply directing people to make the eclipse happen on a different day.

“Seek first to understand.”

Dig in, go see for yourself. Let yourself be surprised by just how hard it is to get stuff done.

 

 

Posted on

Hitting the Numbers by Holding Your Breath

To commemorate the end of WWII, China held a big military parade in Beijing. You can read about it in any number of news sites.

Beijing, though, (as well as Shanghai) is well known for having a serious problem with air pollution.

Here’s my experience: This is a late afternoon photo I took in Beijing in early October 2006… before the pollution was making the international news. Yes, that is the sun. There were days I could not see three blocks from my 6th floor apartment window.

smog

In the run-up to the 2008 Olympics, Chinese officials were determined to present clear skies to the international audience. The same measures were taken again in the run-up to the APEC conference last year. And this article in The Guardian outlines some of the measures taken to present what cynical Beijingers are calling “Parade Blue” skies:

From The Guardian:

Four out of five government vehicles will be taken off the road between now and the parade and private vehicles will be allowed on the roads only on alternate days, based on odd- and even-numbered license plates.

Almost all steel mills in Beijing, Heibei and Tianjin are to be shut down in the lead up to the military parade, Xu Xiangchun, chief analyst at Mysteel Research told Bloomberg last month.

What all of this means is they know what causes the smog. This drill, which was first trialed when I was spending time there prior to the Olympics, is now routine.

OK, so what? And what does this have to do with lean thinking?

We can read these articles and shake our heads. But the story here is that many organizations drive the exact same behavior with their metrics driven cultures.

If making the numbers (or the sky) look good is all that matters, the numbers will look good. As my friend Skip puts it so well, this can be done in one of three ways:

  • Distort the numbers.
  • Distort the process.
  • Change the process (to deliver better results).

The third option is a lot harder than the other two. But it is the only one that works in the long haul. The other two are doing what they did in Beijing – holding your breath.

I’ve seen this manifest in a number of ways. Factory managers doubling down producing product that wasn’t selling to “make absorption” and record paper profits for their plants.

A large corporation that needed good 4th quarter numbers.

  • They shut down production in the last month of the fiscal year to get the inventory numbers up.
  • They pulled orders in from 1st quarter to ship early. This booked revenue as soon as the inventory went into “shipped” status, even if it was still in a container at the port.
  • Because they used LIFO inventory cost models, the deeper they dug into their finished goods, the lower the “costs” and the higher the “margins.” (Don’t confuse LIFO physical inventory management with LIFO cost accounting.)

And they put up some pretty good numbers. Here’s the kicker — they actually believed those numbers, confusing their manipulation of the financial models with actual results. Bonuses all around.

1st quarter, though, was a different story. Shortages resulting in missed shipments because they hadn’t made anything for a month. Thin orders because they had already filled all of the orders they had inventory to fill. (It isn’t like new orders magically appear to replace the ones you ship early.) And at the end of Q1, the CEO had to look the analysts in the eye and explain why.

Anyone can make the numbers for a quarter, maybe even for a year, just as Beijing can clean up their skies for a couple of weeks.

But if you want to make those numbers truthfully, then the only alternative is to “change underlying process.”

This is the difference between a “target” and a “target condition.” To most people a “target” is about the numbers, the desired value of whatever is being measured. That “target” can be achieved by any of the three methods: Distorting the numbers, distorting the process, or actually changing the process.

A target condition describes not only the goal, but how the process should operate to achieve it. The target condition is set by the improver / learner as a next landing point on the climb toward the overall challenge, but the coach hears it  every iteration of the coaching cycle.

Thus the coach (who should be the boss) is well aware of which of the three approaches is being taken (hopefully “change the process”), as well as the issues and problems which must be overcome to get there. It is no longer just a matter of directing that a number be achieved and “holding people accountable” for hitting it.

This two-way conversation is what strengthens the organization and keeps things from becoming silly exercises that achieve nothing but encouraging people to hide the truth. (Then when someone decides to stop hiding it, we have a “whistleblower.”)

Epilogue: As I was working this up, the news has been pouring out about the VW “defeat device” scandal that has now taken out a number of corporate officers, and is going to seriously hurt the company for a long time.

Imagine this conversation in an engineering staff meeting:

Boss: What is the status on reaching the American clean air standards with the TDI engines?

Head Engineer: Don’t worry, we’ve finally got a plan. We are now certain we’ll be able to pass the test.

Boss: Great. Next item.

Contrast that with:

Boss: Where are we on the challenge of reaching the American clean air standards with the TDI engines?

Head Engineer: My current target condition is to pass the test by programming the car to detect when the test is being run, and adjust the engine performance to meet the standards while it is being tested. We’ll met the performance goals by shutting off the emissions controls during normal operation.

Now, that second conversation may well have occurred. But I can see the first case as far more likely.

The question for your organization is which of those conversations do you have? Or is the air too polluted to see clearly?

Posted on

How Do We Deal With Multiple Shifts?

This is a pretty common question.

Today I was talking to a department director in a major regional hospital that is learning Toyota Kata. She picked it up very quickly, and wants to take the learning to the off-shifts.

She (rightly) doesn’t want the night shift to just be deploying what day shift develops, she wants night shift totally involved in making improvements as well. Awesome.

Her question was along the lines of “How do I maintain continuity of the effort across both shifts?” She was jumping into asking how to provide good coaching support, whether there were separate boards, or a single board etc. and playing out the problems with each scenario.

My reply was pretty simple. “I don’t know.”

“What do you want to see your learners doing if they are working the way you envision?”

In other words, “What is your target condition?”

But… how do we coach them, and so on?

I don’t know. But until we understand what we want the improvement process to look like, especially across the shift boundaries, we can’t say. Different target conditions will have different obstacles.

And what worked at Boeing, or Genie, or Kodak, or even another hospital I’ve worked with likely won’t work here in your hospital. The conditions are different. The conditions are different in different departments in the same hospital!

She admitted that she was having a hard time thinking about a target without dealing with all of the potential obstacles first. My suggestion was that this challenge is her improvement board, and the best way to work out a solution was to actually follow the Improvement Kata (that she has been doing such a great job at coaching for the last month).

Trust the process. Once there is a clear target condition for the people doing the work (in this case, the learners / improvers), then we’ll better understand the obstacles we actually have to deal with. That will likely be fewer than every possible problem we can think of right now.

Establish your target condition, then list your obstacles, then start working on them one by one.

The Improvement Kata is exactly the tool to apply when you know you want to do something, but can’t figure out exactly how to do it.

Step by step.

Keep it up, Susan.  Smile

Posted on

Developing Cross Functional Responsibility

The Challenge

It’s a typical staff meeting. The function heads are around the table with the boss. One of them describes a hiccup or problem he is encountering that is outside of his control – it originates in another department for example.

An action item gets assigned in the meeting, and we move on to the next topic.

Good to go, right? Isn’t that the boss’s job?

Let’s expand the role of the boss a bit. Rather than being the conduit of all information, isn’t the role really to “Ensure cross-functional coordination is happening?”

If these meetings are weekly, there is weekly cadence to this kind of coordination, meaning if the issue comes up immediately after a meeting, it is a week before a decision is made. On average, it is a few days.

Let’s look at the nature of the language being used. The implied (but often unstated) question being asked by the function heads is “What do you want me to do?” The even worse implication is “I’ll work on cross functional issues when I get an action item to do it.” Not exactly teamwork.

Here’s another example.

Three functional managers all work in the same building… actually in the same open room. The building isn’t even that big. You can find anyone who is anywhere in the building in less than 5 minutes, just by standing up and walking around.

Their common boss is a in another city, a couple of hours away.

As he talks to these functional managers, they tell him of issues. But they haven’t talked to their counterpart who is 20 feet away. They are expecting the boss to do that. To say this exasperates the boss (who “gets it”) is an understatement.

In yet another organization, we are talking to various department directors about process improvement. Nearly every one of them cited problems in other departments as disruptions to their processes.

These Directors are implicitly (and sometimes explicitly) expecting the CEO and Executive Team to issue directives to the other departments to fix these problems. The problem comes in when the Executive Team accepts the “assignment” and facilitates the communication. Now it’s their job.

Here is the question that surfaced in this organization: The managers were responsible for organizing and managing the processes that are internal to the department. If the Directors aren’t the ones responsible for that cross-departmental coordination… whose job is it? And if it is someone else’s job, what value are the Directors actually adding by managing the managers’ management of the internal processes, and commiserating about the problems from other departments?

All of these cases are the consequence of a management process that sends reports up, and sends decisions down. This develops a deeply rooted unconscious set of habits that are hard to change even when all agree it should be changed.

What Doesn’t Work

Saying “We need to do a better job talking to each other” isn’t going to work. Even saying “You need to talk directly to Dave about that” really doesn’t work because:

  • It is still telling him what to do.
  • The behavior repeats for every instance because “Jim” is still habitually coming to the boss for direction.

What We Are Trying

The objective (challenge) is to get the boss out of job being the sole conduit for cross functional communication. We want these guys working as a team.

In one of these cases, the boss and I took a page from David Marquet’s book, and thought it might help if he (the boss) made it clear that he is going to refuse to be the intermediary in these conversations. Now… how does he create the environment where this cross-coordination is happening as a matter of routine?

David Marquet’s “Ladder of Leadership” model may be useful here.

Grasp the Current Condition

“Start with Awareness, and Just Listen”

         – David Marquet, author “Turn the Ship Around”

imageTake a week and just listen to the words people use when talking about cross-functional problems. Are they simply stating the problem and hoping the boss will pick up doing something about it – and tell someone what to do?

Make a tick-mark on the ladder diagram for the level of each conversation.

Are they implicitly or explicitly looking to be told what to do? (Telling Jim to “Talk to Dave” or even asking “Have you talked to Dave?” is telling them what to do.)

Where is your center of mass?

“Tell me what to do” is the bottom rung. Your own current condition may well be different, but if you have read this far and this still feels relevant to you, it likely isn’t much different.

Establish the Next Target Condition

What words does the boss want to hear when one of those managers is letting him know what is going on? Not in the ideal situation, but at the next level – up one or two rungs.

For example, instead of saying “We’ve got this problem from Dave’s department.” and waiting to be told “Have you talked to Dave?” what does the boss want to hear from this department head?

Maybe “We’ve got this problem from Dave’s department, and I intend to talk to Dave to confirm that he understands what we need from him.”

Apply Rapid Iterations of PDCA

OK, now that the boss knows what words he wants to hear, how does the boss change his response so when he hears “We’ve got this problem from Dave’s department” the boss’s response drives thinking and initiative back down the chain.

Stealing another line from Marquet, maybe the boss says “OK, what do you think I’m thinking right now?”

“ummm… I’m thinking you want me to go talk to Dave about this.”

“Great. What do you expect to happen?” and then “OK, when can you let me know how it actually went, and what you learned?”

Ideally the boss wants to continue this process, setting successive targets until he hears “We had this problem, but Dave and I worked out a solution, and this is what we’ve done.”

or they only come to the boss with a problem that requires the boss to cross-coordinate with one of his peers, but they come with a solid recommendation.

Step by step.

Never give up on your people.

Posted on

Another Homework Question

Another interesting homework question has shown up in the search terms. Let’s break it down:

23. if the slowest effective machine cycle time in a cell is 55 seconds and the total work content is 180 seconds, how many operator(s) should operate the cell so that labor utilization is at 100%?

I find this interesting on a couple of levels.

At a social level, the idea of cutting and pasting a homework question into Google hoping to find the answer is… interesting. Where is the thinking?

What are we teaching?

The question is asking “How many people do we need to run as fast as we can?” (as fast as the slowest machine). But how fast do they need to run? Maybe they only need a part every 95 seconds. If that is true, then I need fewer people, but I am going to run the slowest machine even slower.

In other words, “What is the takt time?” What does the customer need? How often must we provide it?

Then there is the “labor utilization” metric, with a target of 100%. Assuming the planned cycle time is actually 55 seconds (which it shouldn’t be!), we need 3.3 people in this work cell. (180 seconds of labor cycle time / 55 seconds planned cycle time: “How long does it take?” / “How long do you have?” = Minimum Required Capacity)

How about improvement? What do we need to do to get from 3.3 people to 3 people? We can solve for the labor cycle time.  55 seconds of planned cycle time * 3(people) = 165 seconds of total labor. So we need to get that 180 seconds down to a little less than 165 seconds.

Now we have a challenge. We need to save a bit over 15 seconds of cycle time. That might seem daunting. But we don’t have enough information (the current condition) to know where to begin. Then we can establish the next target condition and get started making things better.

These types of questions bother me because they imply all of these things are fixed, and they imply we run “as fast as we can” rather than “as fast as we must.”

Edit: Today I saw two more searches for:

total work content divided by slowest machine cycle time

so it looks like at least two others are working on the same assignment.  🙂

Thoughts?

Posted on

“We Need To…”

When working with large organizations, I frequently hear a surprising level of consensus about what must be done to deal with whatever challenge they are facing.

Everyone, at all levels, will agree on what must be done. They will say “We need to…” followed by statements about exactly the right things, yet nobody actually does it. They just all agree that “we need to.”

I even hear “We need to…” from very senior leaders.

It’s a great car, I wish we made more of them.

– Attributed to Roger Smith, CEO of GM, following a presentation on the Pontiac Fiero.

I can’t come up with a clever name for this, but it is really the opposite of Jerry Harvey’s “Abilene Paradox” where a group embarks on an activity that no one actually wants to carry out. In this case, a group doesn’t take action toward something they all agree must be done.

I would contend that “We need to” spoken to no one in particular is an artificial substitution of the word “we” that does not actually include “I.” Substitute “they” for “we” and you hear what is really being said.

“They need to…”

“Somebody needs to…”

This isn’t clarity. It isn’t accountably. It is a wish.

In Turn the Ship Around, David Marquet challenged (actually ordered) his crew to never use the word “they” to refer to any crewmate on the submarine. This shift in language was an early step toward shifting the teamwork dynamic on the USS Santa Fe. Marquet comments “We don’t have teamwork. We have a rule. You can’t say ‘they’.” but the truth was that the linguistic shift precipitated a shift in the behavior and then the underlying thinking.

This week we asked the question: What small change to their language could we challenge a leadership team to make that would shift the dynamic of “We need to” from general, ambiguous statement toward taking a step to fix it.

What should follow “We need to…” to turn it into accountable language?

One suggestion that came up would be to follow “We need to…” with “…therefore I…”

By making that thinking explicit, we might tacitly flush out “We need to, therefore I intend to wait for someone to tell me to do something.” or “We need to, therefore I am going to hope it happens.” or “We need to, but there’s nothing I can do.”

Realistically, no one would say those complete sentences on purpose, but a struggle to come up with something more concrete might trigger some reflection on the underlying thinking.

Maybe we can turn “We need to, therefore I…” into describing one step the speaker can take in his or her organization without seeking permission*. There is always something that can be done.

This doesn’t need to be scripted or literal. It might just take a self-empowered voice to ask “We all seem to agree on what must be done. What step are we going to take, today, to move in that direction?”

Action Step: Challenge your team when you hear “We need to.” Are you talking about an anonymous “they” or taking a concrete action step? Who, exactly, is “we” if doesn’t include “me”?

Never give up.

_________

*Keeping in mind that “without permission” does not always mean “I have the authority to do it.” It just means “It is the right thing to do, so I’m going to do it.”

Posted on

The Improvement Kata PDCA Cycles Record

The improvement kata has four major steps:

image

Those steps provide a structured pattern to enable consistent practice until they are unconscious and natural.

In the fourth step, “Iterate Toward the Target Condition” we have a form, called the PDCA Cycles Record that provides an additional level of structure for the improver / learner and the coach.

This is the PDCA Record form from Mike Rother’s Improvement Kata Handbook (click the link to go to his download page):

image

The columns in the form correspond with the “5 Questions” that are part of the Coaching Kata.

The intent is that as the coach asks the questions, the learner points to and reads his answers. In the 5 Questions, it is the “Reflection” (on the back of the coaching card) and question #4 that address the PDCA Cycles Record.

Let’s look at how this form structures the learner’s process.

The very first experiment or trial that the learner sets up is based on his understanding of the current condition and the obstacles he is facing. He selects an obstacle, decides what he should do first, and fills that step in Column 1 “Date, step & metric.”

He must think a bit and also fill in “What do you expect?” and describe what effect he expects to have on the process (or what he expects to learn) as a result of taking that step.

Then he hits the yellow bar in the middle of the form. It says “Do a Coaching Cycle.” Do not pass this point without checking in with your coach.

The coach, this time around, is going to ask the 5 Questions, but skip the reflection step, because there is no previous step to reflect on. The coach is (or should be) looking for things like (these are by no means inclusive, rather they just came to mind as I’m writing this):

  • Is the obstacle actually something which must be worked out, or something which must be learned to reach the target? Or is it just a “to do” item? He may ask some follow-on questions to clarify the connection.
  • Is the “Next Step” actually something which addresses the obstacle? Does it reflect a step into “unknown territory” that includes learning?
  • Is the expected outcome a logical consequence of taking the step being proposed? Does it have something to do with the obstacle?

By having the learner write down his intent prior to the coaching cycle, the coach can see how the learner is thinking without biasing that process. He can see if the learner is off track. If so, it’s pretty simple to erase, or even scratch out, the planned experiment and revise during the coaching session.

But either way, as  coach, I want to see the learner’s best effort before I influence or correct it. That is MY process for “grasping the current condition” and even checking the result of a previous experiment on my part by emphasizing something specific during the last coaching cycle.

Once the learner is good-to-go, the NEXT yellow bar says “Conduct the Experiment.” This is the “DO” of PDCA.

Once he is done, the learner is expected to write down his observations in the “What Happened” column, then reflect, and write down what he learned in the “What We Learned” column.

THEN, based on what he learned, plan the next step. So, move down a row, and fill in block #1 with the next step, and block #2 with the expected result.

Then he hits that yellow STOP bar again. This time the coach is going to ask the reflection questions on the back of the card – reviewing the last step and expectation, and then covering the new information: What actually happened; What did you learn; Based on that, what is your next step; and what result do you expect from taking that step?

My job as the coach is to make sure the learner can connect the dots. I want him to write all of that down before I talk to him.

I have to see the learner’s “actual condition now” before I can effectively coach him.

Why Am I Talking About This?

I have run into a few cases now where I have gone into an organization with some prior training or experience with Toyota Kata. They have asked me in to do some additional training, or coach them to the next level because they think they are “stuck.”

In a couple of those cases, I have observed a deliberate* practice of filling out the blocks on the PDCA record during the coaching cycle. Their intent seems to be for the learner to be guided by the coach as he fleshes out what actually happened; what was learned; the next step or experiment; and what is expected and writes those things on the form.

This is very effective if the intent is for the learner to “get it right.”

But from a coaching standpoint, I feel (and this is my opinion) that this practice deprives me of information I need to ascertain how the learner would do it on his own.

I also believe it runs the risk of building a dependency on the coach, and shift the psychological responsibility off the learner – it is easy to fall into the “tell me what to do” trap unless the coach is experienced enough to avoiding “leading the witness” during the coaching cycle.

In most organizations, the hierarchy that likely exists between the coach and the learner has a deeply seated habit of the boss having the answers. I want to avoid reinforcing this dynamic.

A Caveat for Brand New Beginners

When the learner is going through the Toyota Kata steps the first few times, he won’t know what to do. It is completely appropriate for the coach to demonstrate, and guide, the learner through his steps. But the organization should not confuse this effort with the intended pattern of the improvement kata.

As soon as the learner has shown that he understands the intent of the process steps, it is time for the coach to step back and let the learner try it on his own. “Take a few swings” to use a spots metaphor.

That gives the coach the best opportunity to see where he needs to focus his effort. And the PDCA record may well be scratched out, revised, or rewritten in the process. It’s OK for it to be messy. That’s what learning looks like.

_______

*This is different from a case where the learner simply isn’t prepared for the coaching cycle and hasn’t filled in the forms or even thought about what to put on them.

Posted on

How Do You Measure Toyota Kata?

I’ve run into a couple of cases where there is an initiative from a corporate continuous improvement team to “implement Toyota Kata.”

Aside from trying to proscribe each and every step of the way (which runs counter to the entire point of discovering the solution – “you omitted step 7b”), they also expect reports of metrics related to the “implementation.”

Things like:

  • Number of coaches.
  • Number of coaching sessions.
  • Number of active improvement boards.
  • Scoring learners on a dozen or so categories of specific attributes for their coaching session.

Then there are bureaucratic reporting structures demanding that this information (and much, much more) get dutifully filled in and reported up to the continuous improvement office so they could monitor how each site is progressing (often from across an ocean).

I’ve seen this before.

I’ve seen companies try to count kaizen events, and quantify the improvements for each one to justify the payback of the effort.

Similarly, I’ve seen top-level corporate leadership teams struggle to determine how to measure, at a glance, whether a site was “doing lean” to get their results (or getting the results by some other means(?) that were less appropriate).

I’ve seen companies try to manage quality by having the quality staff prepare and submit elaborate monthly reports about what was, and was not getting done based on what they felt was important.

I lump all of this under “management by measurement.” I think it is often a substitute for (1) trusting middle management to tell the truth and (2) actually talking to people.

What’s The Alternative?

First we need to work out what we really want people, especially middle managers, to actually do. What would full participation look like if you saw it?

For example, one company I work with has had a problem where middle managers send their people to internally run “kata training,” then say they have no time to coach their people, give them ambiguous or shallow challenges to work on, and generally look at the training as someone else’s responsibility.

OK, what do you want them to do?

In their book Switch, Chip and Dan Heath talk about “scripting the first moves” so someone who is unsure how to begin doesn’t need to expend limited psychological energy figuring out how to start. We want to get them going. We don’t need to lay out the entire process (that will end up being different for everyone) but we can create some clear rules or guidelines for getting started.

From the Switch Workbook on the Heath Brother’s web site:

Be clear about how how people should act.

This is one of the hardest – and most important – parts of the framework. As a leader, you’re going to be tempted to tell your people things like: “Be more innovative!” “Treat the customer with white-glove service!” “Give better feedback to your people!” But you can’t stop there. Remember the child abuse study [from the book]? Do you think those parents would have changed if the therapists had said, “Be more loving parents!”  Of course not. Look for the behaviors.

Asking that question, in terms of what they would actually see their middle managers doing, we came up with three general actions:

  • Work with their learner / improver to establish a clear challenge for them.
  • Commit to regular coaching cycles with their improver.
  • Commit to receiving 2nd level coaching during those sessions (so they can learn as well).

The question then becomes what mechanism can the organization put into place to encourage that behavior vs. just sending people to the class and not following up in any way.

It is easy to tell people what they shouldn’t do. It’s a little tougher to tell them what they should do.

In this case, we discussed establishing prerequisites for sending someone to the class. But these prerequisites are for the organization sending the participant to the training, rather than just the person attending the class.

It is the sponsoring manager who must commit (perhaps even in writing) to ensuring there is an improvement challenge; to establishing a regular coaching cycle; and to welcoming 2nd level coaching.

Perhaps this would be a commitment to the advance team who becomes a bit of an admissions committee – ensuring the support structure is there before we commit to taking someone on in the class.

We discussed taking a copy of the organization chart and putting dots on it where they had active improvement boards and coaching relationships. That would highlight which organization groups were developing their own people and which ones were doing less of it.

We discussed inviting the middle managers who have sent people to the class but, today, are not supporting, to come to the advanced team with a plan – perhaps a similar commitment – for their renewed participation. But no one would be required, because you can’t force anyone to learn something. Put your energy toward the people who want to learn. Trying to get participation from people who don’t want to do it just frustrates everyone.

And finally, create a mechanism for someone inside a non-participating organization to raise their hand and ask for help with their own development. Don’t punish the entire organization because their boss won’t play.

Now you are talking about mechanics, about actions that have testable outcomes: Experiments that can be set up and tried, improved, and iterated in the direction of something that works for your organization.

It is more work than just measuring people. And it doesn’t work every time. But it works more often than something that never works.

Posted on

Output vs. Takt Time

The team’s challenge is to reach steady output of 180 units per hour.

Their starting condition was about 150 per hour. Their equipment and process is theoretically capable of making the 180 per hour with no problem.

They calculated their takt time (20 seconds) and established a planned cycle time of 17 seconds.

Some time later, they are stuck. Their output has improved to the high 160s, but those last 10-12 units per hour are proving elusive.

This is the point when I saw their coaching cycle.

Looking at their history, they had set a series of target conditions based on output per hour. Their experiments and countermeasures had been focused on reducing stoppages, usually on the order of several minutes.

“Does anybody have a calculator?”

“Divide 3600 seconds by 180, what do you get?”

“20 seconds.”

“Do you agree that if your line could reliably produce one module every 20 seconds that you would have no trouble reaching 180 modules per hour?”

Yes, they agreed.

“So what is stopping you from doing that?”

They showed me the average cycle times for each piece step in the process, and most were at or under 15 seconds. But averages only tell a small part of the story. They don’t show the cumulative effect of short stoppages and delays that can cascade through the entire line.

The team had done a lot of very good work eliminating the longer delays. But now their target condition had to shift to stability around their planned cycle time.

Performance vs Process Metrics

This little exercise shows the difference between a process metric and the performance metric.

Units-per-hour is a performance metric. It is measured after the fact, and tells the cumulative effect of everything going on in the process. In this case, they were able to make a lot of progress just looking at major stoppages..

Stability around the planned cycle time or takt time (you may use different words, that’s OK) is a process metric.

It shows you what is happening right now. THIS unit was just held up for 7 seconds. The next three were OK, then a 10 second delay. It’s those small issues that add up to missing the targeted output.

The team’s next target condition is now to stabilize around their planned cycle time.

Since they averaged their measurements, their next step is to (1) take the base data they used to calculate the averages and pull the individual points back out into a run chart and (2) to get out their stopwatches and go down and actually observe and time what is really going on.

I expect that information to help them clarify their target condition, pick off a source of intermittent delay, and start closing the remaining gap.

Posted on

The Myth of 10,000 Hours

In this TEDx talk, Josh Kaufman, author of The First 20 Hours outlines his theory of learning a new skill.

One of his key points is the prevailing belief that you must spend 10,000 hours practicing a skill to become good at it.

This equates to over 5 years of practice 8 hours a day, 5 days a week.

Intuitively, we all know we have developed competency at tasks in far less time.

The 10,000 hours is practice to become able to perform at an elite, world-class level against intense competition. A top grand master at chess. A winning starting NFL quarterback. To get on the top podium at the Olympic Games.

But as Kaufman points out, the first part of the learning curve – up to basic competency or even above, is very steep. The thousands of hours are for developing on a far more shallow slope into the level of the elites.

In the video, Kaufman broke the basic learning process down to four steps which mirror the structure behind Toyota Kata.

Deconstruct the Skill.

We, in the lean community, made a huge mistake in attempting to deconstruct “continuous improvement.” Instead of deconstructing the process of continuous improvement, we instead deconstructed the production processes that resulted.

We ended up with a list of production process characteristics – “you have to have kanban,” and “you have to have standard work” rather than breaking down the skills that were applied to achieve them.

Toyota Kata gives us a deconstruction of the basics of continuous improvement so that we can learn them. That doesn’t mean there aren’t advanced nuances to learn, but those are built on the foundation of competence with the basics.

And basic competence can get you a long way.

Just to be clear, this doesn’t mean you can get away without understanding the basic tools. It’s just that, in so many cases, they have been put into place without that understanding. The tools are there to help define “what should be happening” and clarify “what is actually happening” so you can apply the correct thought process and use them to drive improvement. Without the underlying thought process, taught by the improvement kata, sustaining the “tools implementation” is very difficult.

Learn enough to self-correct.

Learning enough to self-correct means that you can recognize the difference between what how you are doing it, and how you want to be doing it.

(There’s that “should be happening” and “is actually happening” theme again.)

You can certainly teach yourself to play the piano, or even teach yourself multivariate calculus.There were no flight instructors for Wilbur and Orville Wright, yet they managed to develop very sophisticated skill with aircraft that were dangerously unstable by any modern standard.

But to learn effectively, you need some baseline of what “doing it right” looks like. Sometimes that is obvious. Sometimes it is more subtle.

This is where having a coach really helps. A coach stops you before you get into trouble. He can correct a nuanced imperfection in your technique that may have a major impact. Having a coach can get you to this point much faster.

That’s why many people hire a coach, sometimes by a different name like “taking lessons” or “going to school” to help them get through this phase more quickly.

Most of the good Toyota Kata classroom training out there gets you close to, but usually not quite over, this point. You can try it, you can get through a PDCA cycle with good coaching, but likely you can’t pick up enough to self-correct yet. THAT requires practice with a coach until you catch on to what “good” looks like.

Remove barriers to practice.

If you are coaching or teaching, this is far as you can get someone to go if they really don’t want to participate.The learner has to want to learn the skill, because everything past this point requires internal motivation.

Feeling stupid is a barrier to sitting down and doing the work. The opposite way to say the same thing is “needing to look like you know what you are doing before you know how to do it.”

This is one reason there has to be self-motivation – because you have to be willing to be incompetent in order to build competence.

For some reason corporate training, and related adult education programs, call it “good” if you have your backside in the seat for the required number of hours. Does it matter if you were on your email the whole time? Often no.

So people get sent to the Corporate Training Department’s “Problem Solving” course, get it on their transcript, and are expected to be able to do it. Same thing for “Handling Problem Employees,” “Lock-out / Tag-out” and a myriad of other corporate topics. The goal is almost to simply create a record that you attended, so they can say “We trained him” and remove legal responsibility.

Toyota Kata doesn’t work that way. Many managers struggle with the idea that they have to do more than just attend the class. They are their own barriers to practice.

What are yours?

Practice at least 20 hours.

This isn’t the “myth of needing to practice.” You actually need to do the work.

20 hours of practicing the improvement kata breaks down to roughly 40 days if you are thinking that way on purpose and reflecting on your actions, and consciously self-correcting for about half an hour a day.

The more you try, the more practice you get, and the quicker you will come up the learning curve. It isn’t that hard, but you do have to work at it.

And working at it is more than just reading the five questions.