Home > The Lean Post> Not Every Problem Is a “Nail” But Companies Typically Reach for the Same Old “Hammer”
The Lean Post
Sharing how the world is making things better through lean.

Not Every Problem Is a “Nail” But Companies Typically Reach for the Same Old “Hammer”

by Art Smalley
October 1, 2018

Not Every Problem Is a “Nail” But Companies Typically Reach for the Same Old “Hammer”

by Art Smalley
October 1, 2018 | Comments (14)

Q&A with Art Smalley, Author of Four Types of Problems

When faced with a problem many business leaders and teams mechanically reach for a familiar and standard problem-solving methodology -- whether it’s brainstorming, fishbone diagraming, failure mode effects analysis, value-stream mapping, kaizen events, design of experiments, A3s, 5 whys, 6 sigma, or the 8Ds.

The problem is that the methodology is often mismatched with the problem, creating unnecessary struggle, frustration, delay, and ineffectiveness in solving the problem -- if it is ever solved at all.

In this Q&A with LEI Communications Director Chet Marchwinski, veteran lean practitioner Art Smalley, author of the new book Four Types of Problems: from reactive troubleshooting to creative innovation, explains why settling on a favorite problem-solving technique or two is a mistake.

Most importantly, he explains how to slip the “hammer-and-nail” trap by understanding that most business problems fall into four main categories.

Art was one of the first Americans to work for Toyota Motor Corporation in Japan, learning the principles of its vaunted Toyota Production System (TPS) at the historic Kamigo engine plant, where TPS architect Taiicho Ohno was the founding plant manager. Art learned directly about problem solving from Tomoo Harada, who led the maintenance activities that created the stability that enabled Ohno’s innovations in flow production to succeed.

Q: What is the purpose of the book?  What problem will it help people solve?

Art: Well the problem is that people try to apply one problem-solving approach to all types of problems. But not every problem requires a multi-step A3 approach to find the root cause. Sometimes you just have to contain a problem and get the production line back running. 

The book is about different types of problem-solving routines and why they are necessary. Successful companies like Toyota employ more than one approach in reality. The problem the book addresses is finding a reasonable balance between using a one-size-fits-all approach and a hundred different techniques. I see companies falling into both traps, unfortunately.

Q: The term “lean” doesn’t appear in the title of your book. Was that a purposeful choice?

Art: Definitely. We started off with the working title “Lean Problem Solving,” and I don’t think anyone particularly liked it – especially me. I mean, what is “lean problem solving”? Are we inventing something new here? Not exactly. Problem-solving routines basically derive from the scientific method of plan, do, study, act. It also relies on many specific contributions by different people over the past 100 or so years.

There’s a tendency in the lean management movement to put “lean” in front of everything: lean product development, lean thinking, blah, blah, blah. It gets to be overkill after a while.

During development of the book, we thought about what lean problem solving refers to and said, “What if we talked about several different problem-solving methodologies and also referenced what Toyota does as an example?” After all, Toyota doesn’t problem solve just one way – and yet the average company often falls into this trap.

Q: Who will benefit from reading the book?

Art: Anyone and everyone. I problem solve all day especially with what we call type one routines, which are quick reaction and troubleshooting. It starts at the morning breakfast table with the family and making small adjustments to our daily routine.

"Not every problem requires a multi-step A3 approach to find the root cause. Sometimes you just have to contain a problem and get the production line back running."I practice it in jiujutsu class for example with how to escape from bad positions or submission attempts, etc. At clients, we search for root causes to recurring problems etc. or to attain new target states. I’ve used the four types in service, healthcare, sports, industry, and national laboratories. It should fit everywhere if you grasp the fundamental concepts. 

Q: Can you briefly describe the 4 types of problem solving?

Art: First, you have type one or troubleshooting. In this type, an andon signal is triggered or any type of alarm or concern occurs and an employee responds. If necessary, a supervisor responds too. You try to fix the problem in a few minutes. There’s no A3 report written up or six-sigma team responding. You’re not creating a ton of documentation

So that’s why we have type two or gap-from-standard problem solving, which is a more disciplined and deliberate approach for problems that either recur or are more serious in nature. The key here is that we recognize there’s a gap from a standard in the problem definition, and it persists, or is severe and can’t be ignored. It also can’t be solved by troubleshooting; something deeper and more deliberative must occur. We usually must slow down and focus more to solve it at a more fundamental level. Type two uses a convergent thinking process and extremely critical thinking aimed at cause and effect relationships.

Type three is what in you or I would call kaizen. It’s also called target-state problem solving. So much confusion exists over the Japanese term I thought it was clearer to use the English phrase in order to make people think about it. In this third type, technically we are at a very stable and consistent level of operation so there is no problem, per se. For example, I’m shipping at 100% on-time delivery. Great, so no problem! However, I can still shorten the lead-time through the system to the customer. In sports, for example, the bar is always being raised each season due to competition. Last year’s playbook will not necessarily work as well again this next season. In Toyota, we called this a “created gap” and solving this new problem uses somewhat different thinking skills. It is more divergent and more creative in nature when compared to the previous types.

Then type four is just bigger, better, and even more future-state oriented than the others. It’s commonly called innovation, or we also use the term open ended in the book. It generally spans a longer time frame, involves a bigger change in the product, process, or system, or is entirely new. It starts with a clean slate and reconsiders the fundamental value proposition. Think of breakthroughs in history like penicillin or radar or the internet, etc. and you get the idea.

I like to say that you and your organization must learn to juggle four balls in problem-solving. Mastering just one kind can make you slightly better, but if you can tackle all four effectively as an organization while developing people then you will far outpace the competition.

Q: Why is it important to define a problem type before trying to solve it?

It goes back almost a century to the observation by American philosopher, psychologist, and educational reformer John Dewey that “a problem well-put is half-solved.” Different versions have been restated by various parties over the years.

In general, though individuals and teams simply can’t get traction without a clearly framed problem. Efforts get confused and muddled down as people go in different directions. An important step in the basic process, especially for Type 2 problems, is to clearly define the problem before jumping into the next steps.

Q: Where do popular problem-solving methodologies like A3, 5 whys, 6 sigma fit within the four types – does each methodology correspond to one type?

Most modern popular problem-solving methods like the ones mentioned are some form of Type 2 or gap from standard methodologies.

They can be used in other ways but the original intent for those was to frame problems in a step-by-step manner, get to the root cause (e.g. 5-why technique) and prevent the problem from recurring. Most popular routines involve logic and steps while true six sigma invokes more rigorous statistics. 

The new book Four Types of Problems is now on sale in the LEI bookstore here.

Was this post... Click all that apply
40 people say YES
41 people say YES
25 people say YES
23 people say YES
Related Posts
14 Comments | Post a Comment
Mike Rother October 01, 2018
4 People AGREE with this comment

Hi Art - Yup, there are different kinds of problem solving methodologies and approaches. But, the same scientific-thinking mindset tends to underlie them all. Notice that all four types of problem solving you describe have more or less these same conditions:

• Some kind of goal we are currently not at.

• Obstacles to that goal.

• Solutions that don't yet exist (otherwise they would have been implemented).

• A need to test.

Working on a common underlying mindset makes sense, too, since by definition we can't simply change mindset when moving from one issue or problem to the next. What the Toyota Kata studies found is that many Toyota practices exist to instill a scientific-thinking mindset as a kind of "meta skill," which can then be applied to an infinite range of situations. An understandable error we made was viewing the visible artifacts used in this "training" as the thing itself. When your mentor has you work on an A3 they are not trying to teach you A3, but a scientific way of thinking that you can take anywhere.

Your post seems like a reaction to problem-solving methods getting wielded like a hammer looking for a nail a'la, as you say, every problem triggers an A3 report. I agree it's worthwhile to clean this up, but at the same time we can also go a level deeper and consciously practice and master scientific-thinking skill. Then we, our teams and our organizations can react fittingly to whatever problems or goals we may face in the unpredictable future.

- Mike

Reply »

Art Smalley October 03, 2018
1 Person AGREES with this reply

Hi Mike,

Good to hear from you. Hope we can meet up in person in the not too distant future. I don't really like to exchange messages in this fashion. It often leads to misunderstaning. A couple of quick comments in response: 

1) This is just a short marketing oriented blog post for LEI and not the full contents of the book. I think you will actually appreciate the content if you have time to read it. It never hurts to practice the 3G's as we coach in Toyota. Go and see (read) in this case, get all the facts, and grasp the situation more fully. 

2) I think problem solving (and kaizen, etc.) is a meta skill in Toyota and a big part of the book. Whether you call it a meta-skill or a pillar (i.e. Kaizen and Respect for People)  that has been stated by Toyota for decades. However statements alone don't advance the ball on Monday very well.

3) There is no one Toyota Kata or way to handle problems - I can provide multiple examples from personal experience at Kamigo engine plant alone. This framework outlines 4 different lenses and I could argue for more...but this is a step in the right direction.

4) There are also different instructing, training, coaching, and development routines in Toyota.  Allthough not a thrust of this book it is a topic I hope to elaborate on more fully in the future.

5) Regardelss the current state of problem solving in most organizations is not all that healthy. Even ones practicing A3's, Kata, or Six-Sigma or some other form. They all exhibit similar struggles. So we in Lean still have our work cut out for us.

6) You request in the latter part of your comment to dig a level deeper on the topic of problem solving and skill development. My counter comment back is that the four types and associated routines are a level deeper that has been clearly articulated in the past.  The framework, types, and details for each one goes beyond just abstrate conceptualization or stating everything is "scientific thinking".

No offense intended by any of my comments. I'd like to catch up and speak in person if you have time or by phone.



Reply »

Robert Baird October 02, 2018
1 Person AGREES with this comment

Agree with Mike Rother about developing the complete organization to address all problem types with scientific thinking. Also agree and recommend one scientific problem solving process to be applied in all situations, yes even Type One. These everyday problems addressed as you briefly described, Type One approach could easily be minimized to teams of firefighters treating symptoms. 

I have realized that with consistent use, along with supporting Gemba Walks, scientific problem solving, again at all levels, even Type One problems can go through an A3 within a couple of hours. This is because with developed scientific thinking the A3 becomes truly visual and other people can automatically get involved. 

Reply »

Art Smalley October 03, 2018

Greetings Robert, 

Thanks for the comment. There is a section in the actual book which is a brief history of problem solving influences in the 20th century. It points out that they all derive from the scientific method which is also stated up above in the LEI post.

Where I think people are confused or perhaps naive is the level of detail (scientific thinking in your language) that goes into different problems. Not every problem is equal nor deserves equal amounts of scrutiny. That is just applying the Pareto principle to problem solving. A Toyota assembly plant has 10,000 andon cord pulls in a day - most get handled quickly by a trained troubleshooting response in minutes. Executives also have type 1 responses in their day...Some problems get Type 2 responses and deeper dives.  I don't consider those equal treatments for example.

Even Type 1 routes however do require excellent cause and effect thinking but I personally don't consider that alone scientific. It just gets by on logical analysis and quick actions aimed at mitigation, containment, and return to normal operating conditions. I don't even really consider what I call type 2's or A3's to be the full scientific method. The book elaborates on this if you are interested and so inclined to read it.

Here is an useful example from the news today:

Think of the hole in the ISS space station in as an example...they'll plug it temporarily (type 1) but the root cause (type 2) will take a lot longer to find  Not even a space walk or genba walk will sort that out :-) No doubt better solutions (Type 3 and 4) also exist but those will not be implemented today or this week or this quarter. The level of so called "scientific thinking" is different for each response. The timing, approaches, and actions are importantly different. Think of that as a four types analogy if that helps. 

I actually was invited and worked closely with some of our nations top scientific laboratories for about seven years. I have a healthy level of problem solving experience in Japan with Toyota in addition time working with the strengths and limits of the full blown scientific method in national laboratory environments. That is partly what drove me to write the book

There are a multitude of techniques and approaches required for problem solving, developing people, and fostering involvement in Toyota. It goes beyond hanging up A3's or stating we are scientific thinkers...As you can clearly guess I am not a big "one size fits all" person but everyone is a bit different in that regard based upon their backgrounds, experiences, and learning styles, etc.. Just food for thought. Thanks.


Art Smalley


Reply »

Mike Rother October 03, 2018
1 Person AGREES with this reply

Using the ISS Example:

Target condition: No air leak, right now.

1st Step to test: Put thumb over hole.


Next target condition: Short-term solution, without someone's thumb.

1st Step to test: Epoxy and Kapton tape.


Next Target Condition: Long-term repair.

1st Step to test: TBD


Next Target Condition: Holes like this don't happen.

1st Step to test: TBD


Although the target conditions, complexity and activities differ, the fundamental underlying scientific-thinking pattern is almost always the same. Scientists don’t change *how* they work with each new research topic.

Reply »

Arthur Smalley October 04, 2018

Mike I have a longer response which I assume got to you via e-mail? In the mean time Tom Ehrenfeld has an excellent idea which I hope you will accept. You can call all problem solving "scientific" or "gray" for example and I don't 100% disagree. However I am still going to point out the varying levels of rigor with the types is quite different and not the same exact thinking pattern or management routine. I see four shades of gray where you prefer to see one. The difference is in the definition and details. I hope you'll consider Tom's request. 


Art S.

Reply »

Mike Rother October 04, 2018
1 Person AGREES with this reply

Hi Art - I agree that talking about varying levels of rigor with different types of problem solving makes sense, is useful and is necessary. With Toyota Kata we're exploring what appear to be somewhat universal ("meta") scientific-thinking patterns that underlie good problem solving, and how people can acquire them.

You and I are talking about two different but complimentary topics here. One does not preclude the other. Take a look at the graphic at this link: www.tinyurl.com/TKFindings. I think your topic is in the top part and our topic is in the bottom part, and they work together! - Mike

Reply »

Mike Rother October 04, 2018

Ah, here's an improvement, stemming from this exchange. Cool. It's reciprocal. That is ...

  • Scientific thinking is a foundation that makes different problem solving approaches work.
  • Practicing a problem solving approach can help generate scientific thinking.

I adjusted the graphic accordingly, which you can see here: www.tinyurl.com/TKFindings2. What do you think?

Reply »

Art Smalley October 04, 2018

Hi Mike,

I think we are mostly in agreement at the macro level but where you are comfotable with one depiction of problem solving (one shade of gray) I am uncomfortable with what I often observe in organizations and tend to see a need for 4 shades of gray. Let me try one more analogy...

I can play golf very well with one swing thought and one golf club - a seven iron. Eventually to improve most break the game down into driving, putting, short iron game, long game, etc. measure results and improve.

High level players carry 4 wedges for the short iron game. The swing thought for each differens by situation sand, short grass, long grass, pin position, etc. etc. 

For a beginner the detailed level is too overwhelming. For someone developing skill more specialized clubs, more specialized swing techniques, and different swing thoughts naturally occurs.  

The same is true in Jiu Jutsu (a non Kata based martial art).  Decathalong is one event (athletic endeavor) or it is 10 different events (running, jumping, hurdling, implement throwing, etc.)

You are content to generically refer to it all as scientific thinking and finding the next target condition at the meta level.

I don't find it wrong to adhere to the high level meta skill one way one main thought approch but I *personally* find it less useful and more limiting.  Sometimes you have to dive into the details and look at the individual techniques and approaches under the meta skill. It depends upon the situation, constraints, learner, resources, timing, and other factors.

Most of this I suspect is due to the way we individually like to learn and process information in terms of our learning style. With a survey tool (not part of the book) I can predict who likes a Just Do It approach versus a Generic Step Based Approach versus a true Six Sigma DOE quantitative approach versus a higher level Concept approach.

People can usually identify themselves and where their organization is struggling. Elon Musk an innovator for example is uncomfortable it appears from the outside and less effective in troubleshoot production / logistics hell. 

Hope that make some sense without being too argumentative or confrontational. 

Arthur Smalley October 04, 2018

I forgot to mention one thing that I think is also worth noting in terms of background information...A couple of years ago some retired experts from Toyota in Japan who were developers of the Toyota Way and basic Toyota problem solving approach retired, did some consulting, and then wrote a book.

They wrote the book in Japanese and called it Toyota Style Problem Solving and talked about importance of different approaches as well. It talked about similarities but also differences (Lexus development is not *exactly* the same is fixing a stripped bolt for example in assembly) even though you may want to all generalize it all as scientific thinking. 

John and I wanted to get it translated (I volunteered to do it for free) and published in the U.S. as it nicely depicts differences and types. It was short but sweet. Contractual agreements with the original publisher did not all for this to happen.

So we decided to write our own and actually include a clearer (for us at least) framework, details, and examples for comparison and try to do better. It is most definitely not an attempt to work on the one meta-skill level or one style of inquiry (open ended questions, etc.) by design.

I had already come up with some frameworks on my own. John and I reworked them and came up with the one we included in this book.  I still hope to get the other book translated so people can read other Toyota viewpoints as well.  

In the end they are just that - viewpoints. However I think both books can help many instances where people are stuck and the current approach (not matter what it is) is not working.


Fraser Wilkinson October 11, 2018
1 Person AGREES with this reply

Wow! Art and Mike this is a really great post for the problem solving connoisseur.  As an internal change agent in a large multi-national I am always being asked to help with problem solving so a few years back I tried to explain what Art has described in his visual of the 4 types (Front Line Problem Solving, Amazon).  Even among so called lean or CI professionals I still see people using the hammer all the time so I find the distinction very useful.  In addition, the main mistake I see is mixing up common cause and special cause methods and tools (fishbone to investigate an incident, anyone?).  Also in type 3 some of that gap may be closed by optimisation (you depict a straight line but we know there is variation at work) or by exploiting beneficial special causes. 

It's taken me years to move through the divergent stage of learning many methods to a more convergent stage such as Mike is indicating. 

A few questions, if I may.  I wonder if my journey would have been much quicker if I'd worked at Toyota? Do we need to start this discussion much earlier in people's careers? Is this all a subset of critical thinking skills? 




Mike Rother October 11, 2018

Hi Fraser - I agree, the exchange on this article has been a good one! I like how you say moving through learning to a "more convergent stage" where it's all related to critical thinking skills. That sounds like it mirrors reaching some kind of 'mastery,' where a practitioner can almost effortlessly see what is important now and what should be the next step, in any situation.

I don't think the learning process is particularly fast or overt at Toyota. Maybe it's more that a master/apprentice kind of scientific-thinking skill development is a core, implicit piece of their management approach, and so they keep at it and over time people's thinking moves that way.

I don't think one can, nor should one, bypass the struggle that's involved with acquiring new skill and mindset. It's part of the learning process. Check out this 3-minute video: www.youtube.com/watch?v=dOjs0bn-RDk

On the other hand, I think that, from case to case, it can go faster ... for instance if one is more overt about what you are doing up front, when the practice starts, as you say. We've noticed that Westerners like to know what they are doing and why. Once you explain that you are deliberately practicing scientific-thinking skills in order to get stronger in achieving goals of all sorts – as a team or organization – people often respond with, "OK, let's go." It's pretty cool.

Of course, when practicing any new skill, the learner typically does have to get through the curve shown in the following link. But that's what your coach is there to help you do, as swiftly as possible: www.tinyurl.com/ybouoqv9


Mike Rother October 04, 2018

With regard to "problem solving," of course there are several types. And of course different types of problems can require different approaches. Scientific thinking and problem solving are not the same thing.

What is the difference? What are their respective roles, and how do they interrelate in a work environment? How does that influence what one might do as a manager/coach, or as a consultant? These are things I would ask readers of this exchange to think about, to compliment your taxonomy of four problem types.

Reply »

Art Smalley October 11, 2018

Thanks Mike. I think those are indeed good points. I have a few more posts on this topic which I hope  LEI will run in the near future. I hope they allay some of the potential fears with the four types and provide a mechanism for better coaching as well. I have some examples to share which I think might help clarify some things. 

Reply »

Dueling Methods: 8D and A3
"We Are Unique"
A3 Thinking Roundup
Please include links as plain text URLs only. Do not copy and paste directly from a web page or other document. Doing so may pick up additional HTML that will not function here.
URLs will be converted to functioning links when your comment is displayed on the site.
Here's an example:
See this article for more details: https://www.lean.org/whatslean