Measuring Improvement

One of the most common (and frustrating) problems for the staff lean practitioner is being asked to “measure the savings” resulting from specific improvements.

(This problem is related to, but different from, trying to measure “lean progress” or the status of implementation.)

There are two issues in play here.

First is the level of understanding in the leaders who ask for this in the first place. Frankly, it isn’t their fault. Authors, consultants, and practitioners have been “selling” the concept of “lean production” as a stand-alone thing to do for decades.

It is really easy, in the initial excitement of grasping the potential, to just try to push the tools and promise that great savings will result from simply implementing them.

The initial literature was all about describing the performance of benchmark companies (like Toyota, though there were others), describing the visible tools and saying, in effect, “if you just implement these tools, you’ll get this kind of results.

But making these changes can be expensive. The obvious costs are consulting fees, time (perceived to be) taken away from production work. Therefore, there must be a sufficient ROI to “justify” making the changes.

For the practitioner, the countermeasure is to try to shift the focus to establishing a business objective first.

This shifts the conversation from “justifying improvement” to “what problems must we solve to hit the objective?”

The financial evaluation then shifts from "justifying moving beyond the status-quo” to evaluating alternative solutions to the problem.

If you ask directly, most managers will have things in mind that they would like to do better. The challenge is to get those things framed in enough detail that some value is created for actually getting there.

Why before What

In this TED video, Simon Sinek summarizes a key thing that differentiates an idea that catches on vs. one that plops.

This is relevant to us at a couple of levels.

First, as Sinek points out, truly great companies succeed because they stand for something higher. They have a “why” that drives what they do and how they do it.

Companies that cannot articulate what they stand for are at a competitive disadvantage vs. those who can.

But these concepts are also critical to those of us who are trying to sell the concept of changing the way our own organizations run. Watch the video – then continue below.

In spite of what is taught in the business schools, business decisions are rarely made based on financial analysis and rates of return. Those things are carefully constructed, but often after the fact to justify what someone wants to do already (i.e. has already decided to do).

When we try to sell our changes, we often try to address the “what’s in it for me?” but still continue to try to make logical “what” type arguments.

That doesn’t work. It has to feel right.

Think about your own organization. When or where do things feel like they are going really well, what is aligning? What values are being realized? How do those moments differ from times or places where things are not going so well?

What makes people say “OH Yeah!” ?

As you try to make the case for “lean” or continuous improvement in your organization, are you crystal clear what you believe in? Can you articulate it? Do others in the company want to believe in the same things?

ART: Capturing the Flow in Photographs

I wouldn’t normally post something like this, except for the subject matter: the Toyota assembly line in Valenciennes, France. Stéphane Couturier is a photographer who tries to capture urban and industrial scenes as organic living forms.

Others have found strong metaphors between the Toyota Production System and the organization of natural processes, but these photographs are, well, just cool (to me at least). See the others at this link:

Série "Melting Point" - Usine Toyota n°15 - Valenciennes
2005 - 
C-Print - 160 x 212 cm + marges - 5ex.
C-print - 75,5 x 100 cm + marges - 8 ex.

The artist’s statement seems to have suffered a bit in translation, but the last phrase sums up a Toyota plant:

In front of the Toyota factory assembly line, that is say confronted by a veritable metaphor of movement that is perpetual and implacable as is today’s technological world, rationalized, disembodied, automated and more and more subject to the silent and ruthless profit logic, Stéphane Couturier knows that reality is no longer made up of isolated things, of fixed geometrical shapes, but that it has become a reality of flux, in continuous movement and transformation.

Cool pics, I just wanted to share them.

3D Printing as a Kaizen Tool

One of the tenants of TPS is to learn as much as you can, as quickly as you can, with as much future flexibility as possible. This is the whole point of JIT.

The more quickly something can be built or mocked up, the more quickly it can be tried and tested, and the more quickly we learn what improvements can be made.

We are seeing the beginning of a revolution in fabrication technology as 3D printing starts to move out of high-end prototyping shops and into the mainstream.

This (very entertaining) video tells about an open source(!) 3D printer design that can be had for about a little over a grand. (USD$1250)  (

Open source means that you can grab the technology and scale it if you want to.


Why I Love My 3D Printer

The step from one-off prototyping to full mass-customization is a small one. It is just a matter of time. The ultimate die change is none at all.

So – rather than looking at the limitations of this technology, look at the possibilities.

Coffee + Electrical Panels = 7500

A reader, Josh, sent me this link.

Spilled coffee in 777 cockpit leads to inadvertent hijack warning, FAA-mandated sippy cups look likely

The more compete, technical version, is here.

The short version is:

  1. Airline pilot spills coffee on cockpit panel.
  2. Coffee (or scalded pilot) causes airplane to send out the HIJACK transponder code.
  3. Many people become involved quickly.
  4. Frankfurt bound plane returns diverts to Toronto, passengers are returned (it doesn’t say how) to Chicago, which the author considers a bad thing.


Given that even highly improbable events are nearly certain given enough opportunities, I am actually rather surprised this hasn’t happened already, given the sheer number of opportunities calculated by (#flights x #pilots x #cups of coffee) over, say, a ten year period.

So, given that we have an undesirable outcome (though perhaps not quite as embarrassing as the .45 hole in the cockpit floor from a few years ago), what is the root cause, and what is the countermeasure?

Whatever we do, we should probably have it cost somewhat less than scrambling F-15’s to go ask what the problem is.

(Yes, I am being somewhat serious here, but also struggling just a little to keep a straight face.)

Many companies respond to a similar problem opportunity by banning drinks altogether in the work areas. My guess is that if we wanted to continue to have pilots operate the airplanes, we might consider something else.

f-15-sparrowI will leave my readers to ponder the thought, and remind you that there are worse outcomes than returning to Chicago via Toronto.


Biggest ERP Failures of 2010

pc pointed out a great little article in a post on the discussion forum.

The article touches lightly on why ERP implementations are so hazard prone, and then lists the “Biggest Failures” of 2010.

Of note is that the majority of the listed failures are governments. I can see why. Governments, by their nature, have a harder time concealing the budget over runs, process breakdowns and other failures that are endemic with these implementations.

A corporation can have the same, or even a worse, experience, but we are unlikely to know. They are going to make the best of it, work around it, and make benign sounding declarations such as “the ERP implementation is six months behind schedule” if for no other reason than to protect themselves from shareholders questioning their competence.

Does anybody have any of their own stories to share?

Keep Visual Controls Simple

In this world of laser beams and ultrasonic transducers, we sometimes lost sight of simplicity.

Remember- the simplest solution that works is probably the best. A good visual control should tell the operator, immediately, if a process is going beyond the specified parameters.

Ideally the process would be stopped automatically, however a clear signal to stop, given in time to avoid a more serious problem, is adequate.

So, in that spirit I give you (from Gizmodo) the following example:

Warning Sign

Teaching the Scientific Method on TV

So the Mythbusters are teaching the scientific method as entertainment, and somehow industry is not making the leap that the same thinking applies to management.

Do financial management methods developed by Alfred P. Sloan have such a mass and momentum that there is no way to overcome?

All of the discussions about “change” in the organization really come down to trying to overpower the way business leaders have been taught to think about decision making.

An Open Letter to John Shook

Congratulations on your assumption of leadership at the Lean Enterprise Institute.

The Lean Enterprise Institute has a deservedly unique place in the community of people working to learn and apply the Toyota Production System. The LEI, and the precursor work at M.I.T., by Jim Womack and others, has been largely responsible for moving the Toyota Production System from the domain of a few consultants into mainstream thought. The term “lean,” for better or worse, is now firmly established all in discussions about business management and production efficiency.

From my perspective, though, the LEI is at a crossroads. While what I am offering here is my opinion alone, this platform has given me great opportunities to connect with people across the spectrum of the “lean community.” This is only my opinion, but it was not formed in isolation.


In the early days, the LEI – and Jim Womack especially – took on an almost evangelical role. The initial workbooks – Learning to See, Creating Continuous Flow, Creating Level Pull, Making Materials Flow, had the effect of creating a near canonical definition of “lean.” We could (and did) debate it, but it was pretty clear what the LEI meant by the word.

The message of those early workbooks, carried forward from Lean Thinking, was clear: “We are still learning this, and sharing it with you as we go.”


Today, however, out here in the world of steel-toe shoes, safety glasses (and yes, scrubs and surgical masks), the LEI has developed a reputation for being insular – not particularly open to the idea that people outside of the LEI’s inner circle might have gained deeper knowledge that moves beyond the message of 1999-2001.

In addition, the LEI seems to have drifted from being a standard bearer with a crisp message to becoming a specialty publisher, much like the Productivity Press of the late 1990’s.

If the role of being a specialty publisher is what the LEI is striving to become, then that is great. But it is nothing special and, over time, I feel the brand will dilute and the LEI’s name on a book will mean little more than McMillan, Wiley or Free Press. Even today, I (and others) no longer assume that just because the LEI is publishing a book that it is automatically worth reading or using as a reference.

But the biggest tragedy is lost opportunity to lead.

Today, even though the founders coined and popularized the term, the Lean Enterprise Institute no longer represents the final word of what “lean” means. Nobody does.

Is “lean” a synonym for the Toyota Production System or Toyota’s management system? With all of the reference to Toyota, I would think that is the intent. But there are many well qualified experts out there who make a good case that the LEI’s publications and message are, at best, a subset of TPS. Others argue that “lean” and TPS are fundamentally different in some way.

The result is that “lean” means whatever someone with an opinion says it does.

The LEI has chosen not to engage in this debate. That is well and fine, but at the same time the message emerging from Cambridge is increasingly unfocused – as the shift (drift?) is made from “standard bearer” to “publisher and promoter.”

Therefore, I believe that in the eyes of the world, the LEI no longer owns “lean” and chooses (by action and inaction) to not define it.

Other organizations, such as the AME and SME, are claiming to hold canonical definitions as they work to establish formal “certification” programs to compete with the ASQ’s ownership of “Black Belt.” (I often wonder if Taiichi Ohno would have been able to get a “lean certification” by the standards offered by these organizations.)

The original academic discipline of continuously developing, testing, pushing, revising the current understanding – the very foundations of the LEI out of M.I.T.-  seems to have frozen in place – as though the initial research that led to the LEI’s formation is regarded as the end-all.


So I have some questions.

One of the central tenants of “The Toyota Way” is challenge.

What is the challenge for the LEI? What is that ideal state that the LEI is striving to become?

Is it to be a successful publisher and promoter of books? Or is there still a sense of a higher purpose?

If it is the later, then I would like to suggest that this higher purpose is not felt “out here” in the world.

I, for one, would like to see the LEI return to a position of being the baseline authority. But this means giving up the idea that there is a static understanding which has already been grasped and simply needs to be put into practice. There is simply too much new knowledge being gained for that position to hold.

Ironically, the Toyota management system itself is built on foundation of striving to learn what is not yet understood. It would be appropriate for the LEI to embrace the same thinking. Just as “no problem is a big problem,” I would argue that “We already know” means “we have stopped learning.”


So what would that look like?

Define “Lean” in non-abstract terms.

To regain leadership, the first step would be to describe the target of “lean” as a testable (and refutable) model. When we say “Lean,” it should define what are we striving for.

What does it look like when the organization is truly engaged in continuous improvement? This isn’t about describing tools, or results, but rather the key elements of a system and how they interact with one another – including the process of management and leadership.

What if we put John Shook, Mike Rother, Steve Spear, Jeff Liker, Michael Ballé, Art Smalley in a room – real or virtual – with a skilled facilitator and a task to try and isolate the five or six key operational elements that are in place in an organization that is truly “getting it.” Each of these people brings a significant piece of the puzzle. Could they fit them together?

The idea would be to develop a solid, testable, working theory of what “Lean” means as a general-case model of the Toyota Production Management System. This would be the LEI’s official position. Sure, others have their views, but this is ours, and this is how we are testing it. Put a stake in the ground, then be willing to continuously test it, review it, and update it.

I certainly have my own ideas on what those key elements are are, but would  love to facilitate that conversation and see the results.  🙂

Let’s acknowledge that we learn about TPS by continuous application of the scientific method – research, developing and testing hypotheses, and this is not static knowledge. We have nothing but our best current understanding, and as we apply it, we learn what we do not know. True senseis are those who have mastered, and teach, the process of learning.

Define a process.

Having a theoretical base is not enough. For it to be practical, we also must learn how to put it into practice. We are dealing with two separate things here.

One is the working theory of how TPS works as a management system.

The other is the approach used to deploy that process into an organization that has a long history of not doing it. In this, we must move outside of Toyota’s world and into the very messy world where the people who are practicing these things are isolated from one another, and have little or no coaching or even emotional support.

This means developing another theoretical base – one for deployment.

Lean Thinking outlined a sort-of process with five steps, but I contend those simply describe a sequence of tool deployment and do not address a management system. In those five steps “Pursue perfection” is a core tenant, but has never been well described – at least not until the publication of Toyota Kata. It is time to reflect on the last 15 years, review the approach, and incorporate what has been learned since.

Further, enough companies have tried the sequential-tools-implementation approach over the last couple of decades, that we have a pretty good idea that it doesn’t work very well. But have we truly examined why, beyond the platitudes of blaming “insufficient management commitment?”

Learning to See actually started this process. It outlined a pretty good process for gaining “current condition” understanding. I would certainly be comfortable in saying that if this process does not work to that end, it is more because of flawed application than a flawed process. But subsequent workbooks started down the path of describing the mechanics in ways that largely left out the people and the leadership processes.

So let’s return to the original concept of setting out a path for learning.

Then continuously test

Now comes the tough part. There must be an acknowledgement that, as much as this process captures what we have learned, it is likely imperfect. But we do not know where or how the flaws will show up. They will only become apparent in practice and application.

The only way to widely and continuously test a theoretical base for deployment is to fully engage a population of practitioners. These are the people who are taking the TPS theory and applying it to uncontrolled environments “in the wild” – the actual factories, hospitals, service providers out there.

To be a TPS-based process, the deployment process itself has to have mechanisms for built-in checks, both in application (are we doing it right, if not what is stopping us?) and outcomes (did it work, if not, what actually happened?).

The community of practitioners are going to be the ones on the virtual assembly line, with one hand on the andon cord. How quickly can the community be engaged to respond to issues, swarm the problem, restore to the standard, and then seek to understand what was not previously understood?

What I do know is that the LEI has one of the most active technical discussion forums out there. A webinar attracts thousands of listeners. An email with a link generates a huge spike in hits to the target site. How can that network be organized, focused and harnessed?

I don’t actually know. But if the LEI were to assume a position of leadership and establish a good vision and progressive target conditions to develop the community of practitioners into a team, I am willing to wager that the kaizen process, if properly applied, would work for that as well.

Paradigm shift

This is much more than organizing conferences, webinars, and publishing books written by consultants. Truthfully, anybody can do that, and frankly, there are more than a few who do those things better than the LEI does.

But the fleeting opportunity is one to regain a position of leadership, and truly engage people in ways no one has ever done. But isn’t that the core challenge of the Toyota approach in the first place?