Is This a Problem – Part 2

Last week I posted a story of a failed freezer, ruined food, and a customer support experience that could be summed up as “That’s how we do it.” I invited comments and asked:

“Is this a problem?”

And when I say “problem” I mean, is this a “problem” from the standpoint of the company’s internal process?

There are some interesting comments, some about the internal culture of the company, others about the support process itself.

But I promised to offer my thoughts, so here they are.

The key question is “What did they intend to happen?” While we can speculate, unless we have the process documentation or are otherwise privy to that internal information, we really don’t know what they intended in this case.

Let’s assume, for the sake of argument, that Frank’s experience was exactly as the company intended it to be. Then, from the point of view of their internal process, there is no problem.

“Wait a minute!” I can hear, “Nobody wants  a customer to never buy the product again.”

And here is my point. We don’t know. This company may be perfectly willing to accept that consequence, i.e. “fire the customer” to preserve their warranty cost structure. They certainly would not be the first. Whether that is good business or not is a totally separate issue. The question is “Did they produce this result on purpose, as a logical, foreseeable outcome of the process as they designed it?.” If the answer is “Yes, they did” (and only they can know), then there is no problem. It might be bad business, but the process is working just fine. (I acknowledge that “bad business practices” can result in unintended results – like bankruptcy. But my point is the results are the outcome of a process, and the process is the result of a decision, even if that decision was to “not care.”)

The key point here is that only after there is clarity of what should happen, can the process itself even be addressed. Until the intended result is clear, then there is no way to see if the process works or not.

Was there a problem here? I don’t know. But this is what I would like people to take away from this little story.

Whenever something in your company seems “not right” ask this really powerful clarifying question:

“Did (or would) we do this on purpose?
If the answer is anything other than an unqualified yes then it is likely you have a problem.

Here is a tougher position: If something was unpleasant for your customer, and you don’t intend to fix it, then embrace the truth that you did do it on purpose. Take responsibility for your decisions, look in the mirror, and say “We meant to do it exactly that way, and will do it the same way next time.” If you can’t stomach that, then go back the the first question.

Here is an extra credit question for this little case study in customer support.
What, exactly, did the customer want here?

Amazon.com Gets It

Not many people know that Amazon.com is one of the “places to see” if you are looking for companies practicing the TPS. The fact that their sales and profits are hitting records as most others are scratching and clawing to stay in business is telling.

This recent post by Kevin Kelleher on Gigacom really sums the whole thing up with one sentence quoted from Jeff Bezos’ letter to shareholders:

At a fulfillment center recently, one of our Kaizen experts asked me, “I’m in favor of a clean fulfillment center, but why are you cleaning? Why don’t you eliminate the source of dirt?” I felt like the Karate Kid.

If you have to keep cleaning up a mess, find out where the dirt is coming from.

But the philosophy goes deeper.

If an assembly Team Member is continuously spending time cleaning up threaded holes, go find out how the debris is getting in there (or find a way to keep it out). Go and see.

If you keep losing market share, find out why customers prefer your competitors products. (And don’t sit around a mahogany table talking about it, GO AND SEE.)

Other posts on the same site relate to eBay’s troubles trying to compete with Amazon. The difference, I think, is summed up in a quote from an Amazon executive related to me by someone who was a fly on the wall in one of their meetings:

“At an eBay sellers meeting last quarter, my counterpart was booed off the stage. That is not going to happen here.”

Kaizen is less about the tools than it is an obsessive curiousity about what the next problem is between you and perfection.

How Do You Look At Problems?

A couple of posts ago, I tried to emphasize “hypothesis testing” as the key, core thinking behind the TPS. For that matter, I think that anyone who truly understands any of the various improvement approaches out there will find the same thinking at the core. Certainly Six Sigma; Theory of Constraints; and TQM are all about surfacing and solving problems. They may use different language, might insert the initial lever between different bricks, but in the end, the approaches all embrace the same basic thinking.

I’d like to put out there an idea that it is the way problems are regarded and approached that separates “gets it” from “business as usual.”

What Constitutes “a problem?”

In “traditional thinking” a problem is something which disrupts output. It is something serious enough that it cannot be ignored.

In a true continuous improvement mindset, anything that causes variation from the plan, in any way, is “a problem.” Any barrier between the current condition and the idealized world is “a problem.”

What triggers a response?

In “traditional thinking” if output isn’t disrupted, spend time elsewhere. There is a caveat to this, however. The parable of the “boiling frog” (whether true for actual frogs or not) can drive an ever higher level of numbness as “normalized deviance”   sets in.

Since continuous improvement is a process of discovering the ideal process, variation from the plan is new information. It must be investigated and understood. If everything is running smoothly, then the problem solving shifts to the next barrier to higher performance.

What triggers alarm in the organization?

This one may be the most controversial. While “stopped production” is certainly cause for alarm and immediate response, in the traditional thinking world, it is the only thing that really gets people’s attention.

In a thinking and learning organization, I would add to the above “No problems are apparent.” If there are no andons, there are no defects, there are no line stops, there are no shortages, there are no disruptions, then there is a BIG problem. I say that because these conditions are impossible and it is only because your system is totally numb that you would not see them.

Target Condition

Given the above, then I think it is safe to offer that silence is equated with “stability” in the traditionally reacting organization. Of course it isn’t stable at all, it is just that there is so much systemic anesthesia that nobody feels anything.

In the continuous improvement mindset, things are running as they should if there is a continuous flow of problem being surfaced and solved. That is the only way to be 100% certain that things are getting better every day.

“Management Commitment”

The term “management commitment” is tossed around as a prime reason for failure of improvement initiatives. There are lots of good reasons for this, but until we really define exactly what leaders need to do every day, stop using euphemisms, and start getting real about leadership’s actual role in this process, we are crutching the problem. This is partly “our fault” because we teach the basics very badly. We put top leaders into “kaizen events” but never explicitly link kaizen to daily problem solving. In doing so, we convince them that if only they support enough kaizen events, the organization will be transformed. The logical result is a monthly report on how many kaizen events have been run. Argh.

If we used kaizen events to explicitly teach the core questions, the rules of good process design, and the concept of applying PDCA to everything, we might get more traction. That can be difficult, but maybe if everyone in the industry starts thinking in terms of a few core mantras we might get a chorus going.

The TPS In Four Words

ptolematic_universeIn the world of science, great discoveries simplify our understanding. When Copernicus hypothesized that everything in the universe does not revolve around the Earth, explaining the motions of things in the sky got a lot easier.

In general, I have found that if something requires a great deal of detail to explain the fundamentals, there is probably another layer of simplification possible.

Even today, a lot of authors explain “lean manufacturing” with terms like “a set of tools to reduce waste.” Then they set out trying to describe all of these tools and how they are used. This invariably results in a subset of what the Toyota Production System is all about.

Sometimes this serves authors or consultants who are trying to show how their process “fills in the gaps” – how their product or service covers something that Toyota has left out. If you think about that for a millisecond, it is ridiculous. Toyota is a huge, successful global company. They don’t “leave anything out.” They do everything necessary to run their business. Toyota’s management system, by default, includes everything they do. If we perceive there are “gaps” that must be filled, those gaps are in our understanding, not in the system.

So let me throw this out there for thought. The core of what makes Toyota successful can be expressed in four words:

Management By Hypothesis Testing

I am going to leave rigorous proof to the professional academics, and offer up anecdotal evidence to support my claim.

First, there is nothing new here. Let’s start with W. Edwards Deming.

Management is prediction.

What does Deming mean by that?

I think he means that the process of management is to say “If we do these things, in this way, we expect this result.” What follows is the understanding “If we get a result we didn’t expect, we need to dig in and understand what is happening.”

Control ChartAt its most basic level, the process of statistical process control does exactly that. The chart continuously asks and answers the question “Is this sample what we would expect from this process?” If the answer to that question is “No” then the “special cause” must be investigated and understood.

If the process itself is not “in control” then more must be learned about the process so that it can be made predictable. If there is no attempt to predict the outcome, most of the opportunity to manage and to learn is lost. The organization is just blindly reacting to events.

Here is another quote, attributed to Taiichi Ohno:

Without standards, there can be no kaizen.

Is he saying the same thing as Deming? I think so. To paraphrase, “Until you have established what you expect to do and what you expect to happen when you do it, you cannot improve.” The quote is usually brought up in the context of standard work, but that is a small piece of the concept.

So far all of these things relate to the shop floor, the details. What about the larger concepts?

What is a good business strategy? Is it not a defined method to achieve a desired result? “If we do these things, in this way, at these times, we should see this change in our business results.” The deployment of policy (hoshin planning) is, in turn, multiple layers of similar statements. And each of the hoshins, and the activities associated with them, are hypothesized to sum up to the whole.

The process of reflection (which most companies skip over) compares what was planned with what was actually done and achieved. It is intended to produce a deeper level of learning and understanding. In other words, reflection is the process of examining the experimental results and incorporating what was learned into the working theory of operation, which is then carried forward.

Sales and Operations Planning, when done well, carries the same structure. Given a sales and marketing strategy, given execution of that strategy, given the predicted market conditions, given our counters to competitor’s, we should sell these things at this time. This process carries the unfortunate term “forecasting” as though we are looking at the weather rather than influencing it, but when done well, it is proactive, and there is a deliberate and methodical effort to understand each departure from the original plan and assumptions.

Over Deming’s objections, “performance management” and reviews are a fact of life in today’s corporate environment. If done well, then this activity is not focused on “goals and objectives” but rather plans and outcomes, execution and adjustment. In other words, leadership by PDCA. By contrast, a poor “performance management system” is used to set (and sometimes even “cascade”) goals, but either blurs the distinction between “plans” (which are activities / time) and “goals” which are the intended results… or worse, doesn’t address plans at all. It gets even worse when there are substantial sums of money tied to “hitting the goals” as the organization slips into “management by measurement.” For some reason, when the goals are then achieved by methods which later turn out to be unacceptable, there is a big push on “ethics” but no one ever asks for the plan on “How do you plan to do that?” in advance. In short, when done well, the organization manages its plans and objectives using hypothesis testing. But most, sadly, do not.

Let’s look at another process in “people management” – finding and acquiring skills and talent, in other words, hiring.

In average companies, someone needing to hire someone puts in a “requisition” to Human Resources. HR, in turn, puts that req out into the market by various means. They get back applicants, screen them, and turn a few of them over to the hiring manager to assess. One of them gets hired.

What happens next?

The new guy is often dropped into the job, perhaps with minimal orientation on the administrative policies, etc. of the company, and there is a general expectation that this person is actually not capable of doing the work until some unspecified time has elapsed. Maybe there is a “probation period” but even that, while it may be well defined in terms of time, is rarely defined in terms of criteria beyond “Don’t screw anything up too badly.”

Contrast this with a world-class operation.

The desired outcome is a Team Member who is fully qualified to learn the detailed aspects of the specific job. He has the skills to build upon and need only learn the sequence of application. He has the requisite mental and physical condition to succeed in the work environment and the culture. In any company, any hiring manager would tell you, for sure, this is what they want. So why doesn’t HR deliver it? Because there is no hypothesis testing applied to the hiring process. Thus, the process can never learn except in the case of egregious error.

If we can agree that the above criteria define the “defect free outcome” of hiring, then the hiring process is not complete until this person is delivered to the hiring manager.

Think about the implications of this. It means that HR owns the process of development for the skills, and the mental and physical conditioning required of a successful Team Member. It means that when the Team Member reports to work in Operations, there is an evaluation, not of the person, but of the process of finding, hiring, and training the right person with the right skills and conditioning.

HR’s responsibility is to deliver a fully qualified candidate, not “do the best they can.” And if they can’t hire this person right off the street, then they must have a process to turn the “raw material” into fully qualified candidates. There is no blame, but there are no excuses.

Way back in 1944, the TWI programs applied this same thinking. The last question asked on the Job Relations Card is “Did you accomplish your objective?” The Job Instruction card ends with the famous statement “If the worker hasn’t learned, the teacher hasn’t taught.” In other words, the job breakdown, key points and instruction are a hypothesis: If we break down the job and emphasize these things in this way, the worker will learn it over the application of this method. If it didn’t work, take a look at your teaching process. What didn’t you understand about the work that was required for success?

I could go on, but I have yet to find any process found in any business that could not benefit from this basic premise. Where we fail is where we have:

  • Failed to be explicit about what we were trying to accomplish.
  • Failed to check if we actually accomplished it.
  • Failed to be explicit about what must be done to get there.
  • Did something, but are not sure if it is what we planned.
  • Accepted “problems” and deviation as “normal” rather than an inconsistency with our original thinking (often because there was no original thinking… no attempt to predict).

As countermeasures, when you look at any action or activity, contentiously ask a few questions.

  • What are we trying to get done?
  • How will we know we have done it?
  • What actions will lead to that result?
  • How will we know we have done them as we planned?

And

  • What did we actually do?
  • Why is there a difference between what we planned and what we did?
  • What did we actually accomplish?
  • Why is there a difference between what we expected and what we got?

The short version:

  • What did we expect to do and accomplish?
  • What did we do and get?
  • Why is there a difference?
  • What are we doing about it?
  • What have we learned?

How Strong Is Your Immune System?

Each day you are exposed to an unimaginable number of viruses and bacteria. Any one of them has the potential to overwhelm your body and kill you. But your immune system detects the foreign body, responds, swarms the source of infection, defeats it, and learns so that your immunity is actually strengthened in the process.

Some people, for various reasons, have weak or suppressed immune systems. They suffer from chronic infections. Even if their immune system keeps the infection under control, it is not strong enough to eliminate it. Things which someone else might not even notice take a daily toll on their energy and life.

Your immune system represents your body’s strength to deal with the unanticipated and the unexpected.

In your organization, people encounter unanticipated and unexpected things every day. A small inconsequential defect causes some inconsequential rework. A missing part causes a search and expedite. Each of these things, in turn, propagates more variation, like expanding waves, into the system around it. The variation becomes an infection in your processes, and it has the potential to grow without bounds.

Most organizations have very weak immune systems. They are chronically sick, and expending incredible amounts of time and energy dealing with these “infections.” Because each “infection” is, at best, accommodated rather than eliminated, they tend to accumulate. More and more of the organization’s energy is expended coping.

In the work place this means that the success of the process becomes totally reliant on the vigilance of each individual, working pretty much alone, to see problems and control them enough that some form of output can continue. Worse, many organizations build elaborate systems to assign blame when one of these thousands of opportunities for problems is missed.

Ironically, a lot of “blitz” type kaizen activity actually makes the the system more sensitive to these kinds of problems. If there is no corresponding effort to strengthen the response (swarm, fix, solve) to these problems, it is no wonder that things slide back pretty quickly.

Variation is an infection. And like the viruses and bacteria we are exposed to every day, it will always be there. It is only your ability to quickly detect variation, rapidly contain it, and then deal with its cause that strengthens your organization’s ability to deal with the next one.

Article: Teaching Smart People How To Learn

Greg Eisenbach, in his Grassroots Innovation blog, cites a article that gets to the very root of organizational learning, respect for people, and a myriad of other issues.

The article, Teaching Smart people How To Learn was written by Chris Argyris back in 1991. What struck me about it is that it packs a double-whammy to our “lean” community. Most of us are change agents in some form or fashion, whether with direct operational control, or as either internal or external consultants. The hit comes from the fact that the example dysfunctional organization are consultants themselves.

So in that aspect, we all need to read this and take a long, hard look in the mirror.

The other aspect, though, is that everything here extrapolates to the very organizations we are trying to influence. A couple of key points jumped out at me.

Change has to start at the top because otherwise defensive senior managers are likely to disown any transformation in reasoning patterns coming from below. If professionals or middle managers begin to change the way they reason and act, such changes are likely to appear strange—if not actually dangerous—to those at the top. The result is an unstable situation where senior managers still believe that it is a sign of caring and sensitivity to bypass and cover up difficult issues, while their subordinates see the very same actions as defensive.

I can certainly relate the above from personal experience. It is damned difficult to be open and honest in an environment which does not value openness and honesty!

But then the dilemma hits, because there I am “blaming the client” for my own lack of effectiveness. Instead, it is my responsibility to look at what what I actually did, vs. what I wanted to do; look at my actual results vs. my planned results, and apply scientific thinking. To paraphrase back to 1944, “If the student hasn’t learned, the teacher hasn’t taught.”

The good news is that in the article, Chris Argyris not only points out the problem, he gives an example or two of managers leaders who have overcome it. But they did so only through hard introspection and challenging their only assumptions about themselves, their organizations, and their leadership style.

My last challenge here is this: When we talk about “respect for people” are we talking about behavior which avoids the issues so nobody’s feelings are hurt… or are we talking about being truly respectful and getting the truth out into the open so we can all deal with it?

Management by Measurement vs. a Problem Solving Culture

As I promised, I want to expand on a couple of great points buried in John Shook’s new book Managing to Learn, published by LEI.

A while back I commented on an article, Lean Dilemma: System Principles vs. Management Accounting Controls, in which H. Thomas Johnson points out that

Perhaps what you measure is what you get.
More likely, what you measure is all you get. What you don’t (or can’t) measure is lost.

In his introduction to the book, Shook describes the contrast:

Where the laissez-faire, hands-off manager will content himself to set targets and delegate everything, essentially saying, “I don’t care how you do it, as long as you get the results,” the Toyota manager desperately wants to know how you’ll do it, saying “I want to hear everything about your thinking, tell me about your plans.”

and a little later:

This is a stark contrast to the results-only oriented management-by-numbers approach.

Shook then also references H. Thomas Johnson’s paper. (like minds?)

But I would like to dive a little deeper into the contrast of leadership cultures here.

Let’s say the “management by measurement” leader thinks there is too much working capital tied up in excess inventory.

His countermeasure would be to set a key performance indicator (KPI) of inventory levels, or inventory turns, and “hold people accountable” for hitting their targets.

Since there is little interest expressed in how this is done, the savvy numbers-focused subordinate understands the accounting system and sees that inventory levels are taken at the end of each financial quarter, and those levels are used to generate the report of inventory turns. This is also the number used to report to the shareholders and the SEC.

His response is to take actions necessary to get inventory as low as possible during the week or on the day when that snapshot is taken. It is then a simple matter to take actions necessary. A couple of classics are:

  • Pull forward orders from next quarter, fill and ship them early.
  • Slow down (or even stop) production in the last week or two of the quarter.
  • Shift inventory from “finished goods” to “in transit” to get it off the books.

While, in my opinion (which is all that is), actions like this are at best deceptive, and (when reported as true financial results) possibly bordering on fraud, the truth is that these kinds of things happen all of the time in reputable companies.

So what is the countermeasure?

In a “management by measurement” culture, the leader (if he cares in the first place), would respond to put in additional measurements and rules that, hopefully, constrain the behavior he does not want. He would start measuring inventory levels more often, or take an average. He would measure scheduled vs. actual ship dates. He would measure “linearity” of production.

Fundamentally, he would operate on the belief that, if only he could measure the right things, that he would get the performance he needs, in the way it should be done. “The right measurements produce the right results.”

While not universal, it is also very common for a work environment such as this one to:

  • Attach substantial performance bonuses to “hitting the numbers.”
  • Confuse this with “empowerment” – and perceive a subordinate who truly wants help to develop a good, sound plan as less capable than one who “just gets it done.” He is seen as “high maintenance.” (“Don’t come to me with problems unless you have a solution.”)
  • Look for external factors that excuse not hitting the targets. (Such as an increase in commodity prices.)
  • Take credit for hitting the targets, even when it was caused by external factors. (Such as a drop in commodity prices.)

Overall, there is no real interest in the assessment of why there even is a gap between the current value and the target (why do we need this inventory in the first place?); and there is even less interest in a plan to close the gap, or in understanding if success (or failure) was due to successful execution or just plain luck.

The higher-level leader says he “trusts his people” and as such, is disengaged, uninformed, and worse, is taking no action to develop their capabilities. He has no way to distinguish between the people who “hit the numbers” due to luck and circumstances (or are very skilled at finding external factors to blame) and the ones who apply good thinking, and carry out good plans. Because the negative effects often take time to manifest, this process can actually bias toward someone who can get good short-term results, even at the cost of long-term shareholder value.

This is no way to run a business. A lot of businesses, some of them very reputable, are run exactly this way.

So What’s The Alternative?

Shook describes a patient-yet-relentless leader who is determined to get the results he wants by developing his subordinate. He assigns a challenging task, specifies the approach (the “A3 Problem Solving Process”) then iterates through the learning process – while applying the principle of small steps. At no point does he allow the next step to proceed until the current one is done correctly.

“Do not accept, create, or pass on poor quality.”

He has a standard, and teaches to that standard.

He is skeptical and intently curious – he must be convinced that the current situation is understood.

He must be convinced that the root cause is understood.

He must be convinced that all alternative countermeasures were explored.

He must be convinced that everyone involved has been consulted.

He must be convinced that all necessary countermeasures are deployed – even ones that are unpopular.

He must be convinced that the plan is being tracked during execution, results are checked against expectations, and additional countermeasures are applied to handle any gaps.

And he must be convinced that the results came as an outcome of specific actions taken, not just luck.

In short, even though he might have been able to do it quicker by just telling his subordinate what to do, in the end, that Team Member would only know his boss’s opinion on a particular solution for a specific issue… he would not have taught how to be thorough.

The Learning Countermeasure

If we start in the same place – too much inventory, too few turns – the engaged leader starts the same way, by setting a target.

Then he asks each of his subordinates to come back to him with their plan.

By definition that plan includes details of their understanding of the situation – where the inventory is, why it is. It includes targets – where the effort will be focused, and what results are expected.

The plan includes detailed understanding of the problems (causes) which must be addressed so that the system can operate in a sustainable, stable way, at the reduced inventory levels.

It includes the actions which will be taken – who will do what by when, and the results expected from those actions. It may include other actions considered, but not taken, and why.

It includes a process to track actions, verify results, and apply additional countermeasures when there is a barrier to execution or a gap in the outcome.

The process of making the plan would largely follow the outline in Managing to Learn. The engaged leader is going to challenge the thinking at each step of the process. He is going to push until he is convinced that the Team Member has thoroughly understood – and verified – the current situation, and that the actions will close the gap to the targets.

Rather than assigning a blanket reduction target, the engaged leader might start there, but would allow the Team Members to play off each other in a form of “cap and trade.” The leader’s target needs to get hit, but different sectors may have different challenges. Blanket goals rarely are appropriate as anything but a starting point. But it is only after everyone understands their situation, and works as a team, that they could come up with a system solution that would work.

Of course then the Team Members who had to take on less ambitious targets would get that much more attention and challenge – thus pushing the team to ever higher performance.

Today’s World

Even in companies deploying “lean”, the quality of the deployment is dependent on the person in charge of that piece of the operation. When someone else rotates in, the new leader imposes his vision of how things should be done, and everything changes.

There are, in my view, two nearly universal points of failure here.

  • The company leadership had an expectation to “get lean” but, above that local level, really had no idea what it means… except in terms of performance metrics. This is often wrapped in a facade of “management support.” Thus, there is no expectation that an incoming leader do things in any particular way. (What is your process to “on board” a new leader prior to just turning him loose with your profits and losses?)
  • The outgoing leader may have done the right things in the wrong way – by directing what was to be done vs. guiding people through the process of true understanding.

Fixing this requires the same thinking and the same process as addressing any other problem. Just trying to impose a standard on things like production boards isn’t going to work. The issue is in the thinking, not in the tools.

Conclusion

You get what you measure, but don’t be surprised if people are ingenious in destructive ways in how they get there.

You can’t force a solution by adding even more metrics.

Only by knowing what you did (the process) will you know why you got the results you achieved (or did not achieve). This is a process of prediction, and is the only way people learn.

Learning takes practice. Practice requires humility and a mentor or teacher who can see and correct.

No Blame Requires No Excuses

This little gem is buried on page 54 of John Shook’s new book Managing to Learn, recently published by the Lean Enterprise Institute.

Although it is almost just a passing thought in the overall context, it really gets to the core of a people-supporting culture.

To me, the concept of “No blame requires no excuses” means that the organization has created a culture where excuses are not necessary.

Think about this: What is the purpose of an excuse but an attempt to shift blame from a person to something that person could not control.

So what would it take to remove the need to do this?

First of all, it requires an organizational culture where it is safe to accept responsibility. At that point, excuses are no longer required for survival. Then, and only then, can the team start to deal with the facts as they truly are, rather first working to spin them in a way that is acceptable.

A Firefighting Culture

In honor of October being Fire Prevention Month (at least here in the USA), I’d like to talk about firefighting.

“We have a firefighting culture.”

“We spend all of our time fighting fires.”

We have all heard (and sometimes made) these statements. But I would like to take a couple of minutes and look at what real firefighters do.

They don’t just run in and spray water everywhere in an effort to “do something.”

They study fire. They seek to understand how fires start, how they burn, how they spread. They understand the interactions between fire, air, the specific environment (building structure, outdoor terrain, etc).

They develop a plan to attack the fire. They make themselves reasonably sure that if they do (a), (b), and (c) that the fire will respond in a predictable way.

They execute the plan.

They watch the results. If the fire behaves the way they predicted it would, they continue with the plan. If something unexpected happens, they pause their thinking long enough to understand what additional factor may be at play; what they might not have known or considered. They seek to understand the situation whenever something is going differently than what they predicted.

They adapt the plan to account for the new information or the changed circumstances.

They continue to do this until the situation is under control, and the fire is out.

While doing these things, they work methodically. They verify success at each step of the plan – they do not move ahead of their confirmed progress (which would put fire behind them and block their escape route).

While theirs is a dangerous business, they do not, as a matter of course, put human life in jeopardy simply to save property. Heroics are reserved for saving the lives of others.

Once the fire is out, the fire investigators arrive. They seek to understand how this fire started. Where did fuel, oxygen and a source of ignition come together and how? What fire suppression mechanisms failed, and why? How, why, did it spread? Did containment fail? This information is incorporated into the knowledge base of the society in general in the form of regulations, building codes, electrical codes – countermeasures.

In short – firefighters relentlessly follow PDCA.

Now, I must admit that, occasionally, in the excitement of the moment, firefighters get ahead of themselves, or rush into something they don’t fully understand. We usually know when this has happened because there are somber processions and bagpipe music. But even then, they seek to understand what happened, why, and improve their process to prevent recurrence.

Now, the next time you say “All we do is fight fires” consider the above. My guess is that you aren’t fighting anything. Rather, you are running around, making a lot of noise, but tomorrow the building is still burning – just in a different place.

Don’t forget to check your smoke alarms and change the batteries!

Be sure to read the follow-on post here: /2011/01/17/firefighting-kata/

Bloodletting: Why Controlled Experiments are Important

Bloodletting: Why Controlled Experiments are Important

I want to start this post with the last paragraph of this article:

Next time someone tells you that they are sure their idea will work, consider running a prototype in a controlled experiment, and make a data driven decision!

Now – the article itself is in the context of Microsoft software development, but this is what "lean thinking" is all about.

In the Toyota Production System, processes and the organization are deliberately constructed so that a kaizen experiment in one area is unlikely to affect another – at least not before there is ample warning and opportunity to reverse the change.

Every production cycle itself is conducted as a controlled experiment testing the hypothesis that the process:

  • Can be conducted as specified.
  • Delivers the specified results.

Of course – and here is the key point – it is only an experiment if someone actually examines what really happens. And this is the fundamental difference that is captured in the quote at the top.

A truly lean process has built in checks that check, each and every time, whether the idea works. Any time an idea doesn’t work, there is an expectation that the process will be stopped and understood.

Continuing to blindly produce, without knowing if each and every process is working as planned, without knowing if the result is as planed each and every time is being sure it will work. It is bloodletting.

Don’t bleed your process, or your company, to death with things you "know will work."