Four Types of Problems

Most problem-solving training teaches a single method and implies it works everywhere. In practice, the hard part usually comes earlier: recognizing what kind of problem you are looking at. A blown fuse on a machine, a quality defect that keeps recurring, a cost target that demands a better process, and the search for a product that does not yet exist are four different problems. They call for four different kinds of thinking. Treating them all as one is the most common — and most expensive — mistake in problem solving.

1Background

Problem solving may be the most fundamental of organizational skills. An organization with only pockets of problem-solving ability struggles over time; an organization with an army of capable problem solvers is far better equipped to face whatever the market brings. Yet despite shelves of books and decades of training programs, problem solving as a discipline often seems stuck — or slipping backward. People reach for a familiar tool before they have understood the problem, mistake activity for progress, and declare victory at containment.

The four routines described here reflect patterns of problem solving I observed at Toyota's Kamigo engine plant during my time with the company, and in the years since. They are not a method Toyota hands out on a card; they are a way of organizing what effective problem solvers actually do.

A common question at Toyota is "What problem are you trying to solve?" On the surface it sounds simple. What it really asks is deeper: what type of problem is this, and how will you proceed? The answer is not one size fits all. Before you can solve a problem well, you have to recognize what kind of problem it is — and the four types are a way of answering that before you pick up a tool.

A useful analogy is a golfer's bag. A professional is allowed fourteen clubs, and the skill is not only knowing how to swing but knowing which club the situation calls for — a different shot for the wind, the lie, the distance, the hazard. Problem solving is the same. Mastery is not one method executed forcefully; it is recognizing the situation and reaching for the approach that fits.

2A short history of modern problem solving

The four types did not appear from nowhere. They sit at the end of a long line of thinking that runs across the whole twentieth century and two continents. Knowing that lineage matters, because it reveals something reassuring: beneath the competing brand names, there are far more similarities than differences. Nearly every method is a descendant of the scientific method — observe, define, hypothesize, test, check, learn.

The intellectual roots run from the scientific method and critical thinking into the early twentieth century. John Dewey, in How We Think (1910), set out a reflective-thinking sequence — define the problem, analyze it, propose solutions, evaluate them, select one — that most modern methods still echo. Frederick Taylor separated planning from doing. Walter Shewhart, a physicist at Western Electric, built the control chart in 1924 and, in 1939, described a specification–production–inspection cycle that turned inspection into a learning loop. Ronald Fisher formalized the design of experiments; Vilfredo Pareto's 80/20 observation later became Juran's Pareto Principle.

After the Second World War this thinking reached Japan through several channels at once. Homer Sarasohn and Charles Protzman taught a five-step problem-solving model to Japanese executives. W. Edwards Deming brought Shewhart's statistical methods and the product-focused cycle the Japanese named the Deming Wheel. Joseph Juran arrived in 1954 emphasizing quality as a management responsibility. The Union of Japanese Scientists and Engineers (JUSE), and Kaoru Ishikawa in particular, codified much of what we now think of as Japanese problem solving — the PDCA cycle, the seven QC tools, and shop-floor QC-circle activity.

One point deserves care, because it is widely misstated. Toyota did not simply adopt what JUSE taught. Shigeru Mizuno and other academics visited Toyota under the JUSE banner and offered advice; what Toyota built afterward was its own internal version, adapted to its own work. The same pattern shows up across the company's history — Toyota took outside ideas, dismantled them, and rebuilt them as something of its own. By the early 1960s, Toyota and other Japanese firms were using compact six-step problem-solving routines, alongside more elaborate twelve-step formats for QC circles. Those step counts were never perfectly standardized, but they all carried the same underlying logic.

Lineage (illustrative)	Steps
Dewey — reflective thinking 1910	Define → Analyze → Propose → Evaluate → Select
Sarasohn & Protzman postwar Japan, 5-step	Define clearly → Get the facts → Analyze → Put plan into effect → Monitor & adjust
TWI Problem Solving 1955, 4-step	Isolate → Prepare → Correct → Check & evaluate
Toyota-era six-step early 1960s	Define → Set goal → Identify root causes → Implement countermeasures → Check → Follow up & standardize
QC-circle twelve-step 1960s, shop floor	From identifying and selecting problems through analysis, causes, solutions, trial, implementation, confirmation, and follow-up

Later branches added to the toolkit without changing the core: the U.S. Department of Defense's 8D (popularized by Ford), Six Sigma's DMAIC, the A3 report at Toyota and Bridgestone, design thinking, FMEA, and the Russian inventive-problem method TRIZ. Each suits a particular kind of problem. The four types are a way of organizing that abundance — not adding a hundred-and-first method, but giving you a way to decide which of the existing ones the situation actually calls for.

3Type 1 — Troubleshooting

Troubleshooting is reactive problem solving aimed at the immediate symptom. Something has gone wrong and needs a fast response: a machine has stopped, a line is down, a shipment is late. The goal is relief — restore function now. Troubleshooting is necessary and valuable; an organization that cannot respond quickly to abnormality will not survive.

It is also the type most people overestimate. Almost everyone believes they are already good at it; very few are. Real skill — reading a situation fast, ruling out the unlikely, restoring function safely and without making things worse — takes years to build. It is the daily craft of first responders, emergency medical teams, and the military, where a slow or wrong reaction is measured in lives. On a factory floor the stakes are usually lower, but the skill is the same, and just as rare.

Toyota's command of Type 1 is one of the most underappreciated strengths of its system. A large Toyota plant sees several thousand andon pulls a day — each one a small problem surfaced the instant it occurs and resolved on the spot, in seconds or minutes, by the team member and team leader closest to the work. That does not happen by accident. It is built through years of training and through the structures standing behind it: the Floor Management Development System that keeps a trained team leader within reach of every station, and the three-pillar activity and daily management that turn rapid response into a disciplined, visible habit rather than improvisation. It is jidoka in daily practice. Most organizations admire Toyota's root-cause problem solving and miss this: the sheer volume and speed of competent troubleshooting underneath it is itself a massive, hard-won advantage.

Its limit is also its definition. Because it treats the symptom, troubleshooting alone does not reach the underlying cause, and the problem returns. Relied on by itself it becomes a cycle of firefighting. Done well, though, it buys the time and stability to do Type 2 properly — and at Toyota's scale, doing it well is a competitive weapon in its own right.

Typical steps — the 4C's^*

1. Concern — what is the concern? (for example, the defect or abnormality observed)
2. Cause — what is the immediate cause?
3. Countermeasure — what action restores normal operation?
4. Check — did the countermeasure work?

^* In Four Types of Problems I describe Type 1 with this simple 4C's framework, drawn from the kind of defect-tracking templates long used on the Toyota shop floor. It is deliberately fast and lightweight — the point of troubleshooting is speed, not depth.

4Type 2 — Gap from Standard

Gap-from-standard is structured problem solving aimed at preventing recurrence. A standard exists, performance has fallen short of it, and the aim is to close the gap by eliminating the underlying cause so the problem does not come back. This is the territory of clear problem definition, goal setting, root-cause analysis, countermeasures, confirmation of results, and standardization — the disciplined core most people picture when they hear "problem solving."

Type 2 is not a single method. Several structured approaches all serve the same gap-from-standard purpose: the eight-disciplines (8D) method common in the automotive supply base, the broad family of practical problem solving, and Toyota's own eight-step method, often called Toyota Business Practices (TBP). They differ in detail and vocabulary but share the same spine — define, analyze, find the cause, counter it, confirm, and standardize. It is the deepest and most widely used of the four types, and it will get its own full treatment.

Typical steps — Toyota's eight-step method (TBP) as one example^*

1. Clarify the problem.
2. Break down the problem.
3. Set a target.
4. Analyze the root cause.
5. Develop countermeasures.
6. See countermeasures through.
7. Evaluate both results and process.
8. Standardize successes and share the learning.

^* Shown as a representative example of Toyota's current eight-step method. Other Type 2 approaches — 8D, the wider practical problem solving family, and the earlier Toyota-era five- and six-step routines — order and name the steps differently while following the same underlying logic. Toyota's method has itself changed over time; this is the present-day form, not a fixed definition of Type 2.

5Type 3 — Target Condition

Target-condition problem solving raises performance above the current standard, even when nothing is "wrong." Today's process may be meeting its goals and still be full of waste, overburden, and unevenness. Type 3 asks a different question from Types 1 and 2: not "why did performance fall?" but "how should this ideally work?" It pursues a should-be condition — shorter lead time, better flow, higher built-in quality — through divergent, creative thinking rather than convergent root-cause analysis.

In my book I call this type target condition (or target state). In everyday lean language it is essentially kaizen — continuous improvement beyond the standard. Because it is such a large subject in its own right, it already has a full canonical guide on this site.

Typical steps — the six-step kaizen cycle^*

1. Discover the improvement potential.
2. Analyze the current situation.
3. Generate original ideas.
4. Develop the implementation plan.
5. Implement the plan.
6. Evaluate the new method, then begin again.

^* Shown as one representative version. There are many forms of the kaizen cycle — six, eight, and twelve steps among them — and the right one depends on the topic (work-method, flow, or equipment kaizen). The Kaizen guide works through this six-step version in depth.

6Type 4 — Open-ended

Open-ended problem solving is innovation — creativity, synthesis, and the recognition of opportunity. There is often no defined problem at the start. Instead of correcting or improving an existing thing, Type 4 identifies new problems, new solution spaces, or new value that was not previously recognized: a new product, a new process, a fundamentally different way of doing the work. It achieves radical rather than incremental improvement, and it tolerates — even expects — failure as part of learning.

Type 4 is usually associated with product development, but any function can be innovated. The same cutting-chip problem that began as a Type 1 nuisance and a Type 2 root-cause case eventually drove Type 4 leaps at Toyota — new sensing technology, redesigned washing methods, and transfer-machine architecture that changed the problem entirely.

This is the type that matters most for the long run. Types 1 through 3 keep a company in business — they restore, stabilize, and improve what already exists. Type 4 is what creates new business: new products, services, and processes, and the growth and opportunity that come with them. Without innovation, an organization cannot compete indefinitely, no matter how well it executes the other three.

Typical steps — the Stanford d.school design-thinking process^*

1. Empathize — understand the people and their needs.
2. Define — frame the right problem to solve.
3. Ideate — generate a wide range of possibilities.
4. Prototype — build something testable, quickly.
5. Test — learn from real feedback and iterate.

^* The design-thinking process from the Hasso Plattner Institute of Design at Stanford (the d.school) is shown as one well-known example. It is one of many innovation frameworks in use — TRIZ, lead-user methods, and others serve the same open-ended purpose. Unlike Types 1–3, open-ended work has no single canonical sequence.

7Conclusion

The field of problem solving is, historically, well established. In practice, it is rarely mastered. Most organizations — even now — struggle with the basics, and much of that struggle begins before any method is applied, in how the problem is seen and framed. Sorting the competing approaches can be daunting. One method will not fit every case; but the answer is not to collect a hundred methods either.

The four types are meant to cut through that. They are not a new technique to add to the pile — they are a way to recognize, quickly and honestly, what kind of problem is in front of you, so you can reach for the right approach and apply it well. Troubleshooting to restore function. Gap-from-standard to eliminate recurrence. Target condition to raise the bar. Open-ended to create what does not yet exist. An organization that can do all four — and knows which it is doing — can attain stability, sustain its gains, and keep advancing toward its goals. To paraphrase Juran: it is better to do a few things well than a multitude poorly.