The Days at General Motors
The seedlings of what would later become the Breakthrough Strategy were planted in Dr. Harry's mind when he was recruited into a General Motors factory in Anderson, Indiana. The year was 1976, the division was Delco-Remy, the operation was Plant 3 where GM produced engine starter motors. Dr. Harry's first role at GM was to supervise one of the final assembly operations. While working at the plant, Dr. Harry recalls a program called "TQ," or Total Quality. Some nicknamed the initiative "Think Quality" as a way of keeping the idea in the forefront of their minds and actions. In reality, quality was more of a mantra than it was a practice at that time, as was the case for most American companies before and during the decade of the seventies.
The Strange Idea of Quality
With the notion of quality hanging in the background, GM was speaking the language of business. The factories and their leaders were judged by their adherence to production schedules, reduction of scrap and realization of cost targets. From an operations point of view, it was production yield they were after. In this context, yield was one of the principal measures for gauging the relative efficacy of an operation. As a manufacturing manager, your rewards were largely based on production yield, compliance and cost.
If you were a high flyer, these were the preoccupying factors by which you would breathe, eat, live or die. And oh, by the way, the quality department might point out that you had a problem with your solenoids, armatures or field coils - or with this, that or the other. While your lips were saying "Oh yeah, I'll get that -fixed," your heart was saying "only if it doesn't compromise my churn rate."
After all, why hold back a product that can be shipped today simply due to a small quality problem, knowing the warrantee program will fix it later? This was the predominant thinking and practice of the day. Dr. Harry remembers all too well a certain senior manager saying "You must always remember that we build Chevrolets around here, not Rolls Royces." While the system loosely reinforced the idea of quality, hard rewards were doled out on the basis of quotas that were blind to that idea.
Quality was an issue to "get around" or otherwise circumvent, and production managers considered the QC department to be more of an adversary than a helping hand. It was an organizational entity with which you often had to negotiate the reduction of defects so as to get product through the system. In this sense, the basic mission of the quality control department was little more than a negative distraction from your principal mission to ship as much product as possible. The quality function could only hinder you, not help you, in meeting your personal goals, which were all built around meeting schedules and budgets.
At the time, no one could imagine in their wildest dreams shutting down a final assembly line in Detroit because of a quality problem at a local factory in Indiana. Keep the line running; if the product is bad, don't worry about it; we have people at the end of the line who can repair it. The very last thing a production manager wanted was to shut down an engine plant, or even try to fix problems in line, because the perception was that the consequential risk of doing so was enormous. Quality control was something to be dealt with at the end of the line, not during the course of production.
In this sense, it was yield of schedule that mattered, not yield of quality - not the yield of "good" product. The whole idea behind mass production was to keep the line running, preserve he income stream and keep pace with rising demand. The concern was only with the amount of product required and the cost of achieving the desired quantity - no more and no less.
At any time, we only know what we know and, perhaps more importantly, we don't know what we don't know. Back then, no one conceived that a three to four sigma product could cost a corporation as much as 30 percent of sales. One or two percent was believable, but not 30 percent. Now with the huge benefit of time, we understand that giving up some production time is often found to be a highly viable trade off to get a pervasive quality problem fixed. But that's what we know now.
Apparent to Dr. Harry in 1976 was the army of people on the floor who regularly came in contact with quality problems but who, at the same time, had no incentive for recognizing the problems, no idea how to solve them and no motivation to do so. They didn't have the knowledge, they didn't have the tools and they certainly didn't have the leadership or management backing. Something was definitely wrong, particularly given the fact that there were people who could fix the problems on an ongoing basis, and teach others to fix them, if only empowered to do so.
That was Dr. Harry's observation as he watched a sea of skid boards form - enormous loads of copper field coils ready to be scrapped in the name of keeping up with production, minimizing chargeable inventory and ensuring good housekeeping.
There was a quality control department, but its primary role was to perform on-line and off-line test and inspection at the end-of-line point, as well as "red-tag" rejects and administer the accompanying paperwork. Sweetheart deals with quality control management were not unusual; they would release "slightly defective" product in the interest of helping production managers meet end-of-month product shipment schedules. When such deals were made, the QC manager would then have a bank of favors he could leverage to eventually get some quality goals accomplished.
In short, quality was circumstantially negotiated, and the more severe a problem was, the greater power the QC department had in such negotiations. Clearly, there was a need for a hard methodology. Dr. Harry's experience at the Anderson plant led him to the distinct opinion that, somewhere, there must be and unbiased science for solving quality-related problems, not only in manufacturing but in the administrative departments as well. While such problems were not as visible as the skid boards of factory waste, there were loads of mistakes in accounting, in labor relations, in all manners of paperwork.
Driven by his curiosity, and his ambition, Dr. Harry began to intensely study the work of Dr. Deming and Dr. Juran, as well as several of the leading Japanese quality consultants. He read about theory X, theory Y and theory Z. He learned the Kepner-Tregoe philosophy and method of problem solving, and he became familiar with various other problem-solving methods designed to rapidly discover where the problem is and where it is not. "Total Quota," he used to call TQ at GM. While there was obvious merit to meeting the production schedule, there was a serious flaw in thinking you could maximize capacity without improving first-time yield by solving chronic quality problems.
As he explored, Dr. Harry came to the fundamental conclusion that quality could and should be managed, the idea of which he clearly derived from Drs. Deming and Juran. A related conclusion was that any effective method for managing quality could work hand-in-hand with the quota-driven thinking he encountered at GM. Another vital conclusion Dr. Harry made during his early years at GM was that a corporation could greatly increase the quantity and quality of problem-solving, and could improve cycle time, by engaging in a systematic program of knowledge transfer. Here again, sitting on the other side of nearly 30 years in industry, we can readily see the wisdom of knowledge management in a global economy. But this wasn't so obvious to those who were enveloped by a system that rewarded quotas, not quality.
The idea of methodically fixing and preventing problems in real time through a data-driven approach, coupled with the idea of transferring that knowledge to a wider base of people, is what began to form in Dr. Harry's mind as quality leadership. It's not enough that we have the vision and methods for solving problems. That's managing for quality. We must proliferate that vision and the enabling methods throughout a corporation, and empower people to keep the idea of quality alive. That's leading for quality. It all seems so logical now, but at the time this thinking was more like heresy than logic. When Dr. Harry would share his ideas, many of his colleagues would literally think he was nuts. "That sounds good in theory," they would say. "But we don't have time for that. We have a factory to run. You do want to get your paycheck, don't you? You keep talking like that and you'll never get promoted."
Observations of a Developing Leader
Maybe Dr. Harry was more concerned about new ideas and innovation than he was about a paycheck. In questioning the status quo, he provided a window into what motivates a leader. If Six Sigma is about leadership - and it most, most certainly is - then those who would be Champions or Black Belts must be ready, willing and able to cut against the grain with every fiber of their being.
For the next several years, Dr. Harry progressed in responsibility at GM. He changed departments, worked multiple shifts and was involved in operations from the raw material stage to hard manufacturing to final assembly. He managed machine shops that built tools, dies and fixtures. He interfaced with customers and different plants. As he rotated through the system, he was continually struck in the head with the holes, inconsistencies and downright foolishness he encountered.
When working in a large tool shop, Dr. Harry took note that many engineering drawings arrived flawed and needed to be "red-lined" for corrections. Rather than focusing on the yield (quality) of its design and engineering drawings, the focus was on correcting faulty features and specifications during production. An industrial-era operation was relying on its highly experienced machinists and toolmakers (craftsmen) to correct systemic problems on the fly. In effect, a pre-scientific mindset was guiding modern industrial practice. There was a 100-year-old leftover in the back of the refrigerator, and no one had the time or wherewithal to throw it in the trash, then replace it with fresh, high-quality intellectual food.
At the same time, there were highly automated, refined, reliable operations in different areas of GM. The bell curve applied. On average, the company was operating according to the principles of science, but at the extremes there were some very antiquated, and very advanced, practices. GM was a proxy for other corporations of its day in that it inconsistently embodied different economic eras in real time and space.
It was clear to Dr. Harry that his company not only had to change the primary signal of its thinking, but it also had to narrow the obscuring noise around that signal. It was also clear that the way to do this was to rid the company of its leftover craftsman-era mindset. There was no place in the machine age for fuzzy philosophical thinking and fix-it-after-the-fact practice. If you want to change the signal and decrease the noise, you need constancy of leadership based on the hardboard of science.
For Dr. Harry, this meant nothing less than a standardized problem-solving methodology for all to follow. It meant building such a method on the basis of hard facts, data and artifactual evidence. It meant driving and instituting such a method from the top of a corporation, and providing the support and tools required to make it work. It meant training people and sharing knowledge, consistently and systematically.
At the time, Dr. Harry remembers thinking the world needed a highly repeatable system for realizing the practical aims of TQM, not a TQM philosophy. It needed to change the signal and decrease the noise of how quality management principles were applied and integrated into the fabric of business. It needed to make TQM more like a science and less like a rally call to "do it right the first time" or "continuously improve." At the same time, the deflating truth was that Dr. Harry's lone vision in the woods of GM had little potential for shifting the tide of quality thinking and practice. Only a few years into his career, his chance was about a snowball's in hell. Besides, he realized his own shortcomings in terms of knowledge and experience.
Nevertheless, true to his nature, Dr. Harry continued to scrutinize and question the way his company conducted business - particularly at the process and operations levels where he was most involved. As we mentioned before, he had studied many problem-solving methods and had spent a great deal of time thinking about how the scientific method could apply to improving process capability and operational performance. He continued to read and experiment with new ideas and methods, such as statistical process control methods and design of experiments.
Through direct observation, Dr. Harry simply knew there had to be a better way. He recalls a time when he became involved in a quality problem that posed a risk of compromising the functionality of a certain type of starter motor. Small wire coils, called field coils, were wrapped around an armature to set an electrical field. During production, the coils were dipped in a fluid that became plastic after it cured. Sometimes, the plastic coating would develop pin holes, thereby increasing the potential for undesirable electrical grounding.
Dr. Harry designed an off-line, rudimentary factorial experiment to isolate the variables that were responsible for the pinhole problem. Even though the experiment was successful, several of those involved said the success was more due to luck than science - only because they didn't understand the analytical logic and inner workings of a factorial experiment. As satisfying as it was, Dr. Harry's success highlighted a much more pervasive and important problem. His company, and many others, still didn't have a unified, singular approach for solving problems that everyone followed and wholeheartedly trusted. All the existing approaches had merit, but not one was leveraged synergistically within and across the many organizations of a corporation.
Symptomatic versus problematic thinking.
At the time, Juran had a method for solving problems, as did the Kepner Tregoe firm and others. But their methods were clearly slanted toward SPC charts, defect diagrams, brainstorming and systematic trial-and-error approaches for finding solutions. It was a time in which the quest for solutions was driven by symptomatic thinking. If the field coil coating has pinholes, then we better look for the causes of that symptom, not a better way to repair the holes (as was the case at GM). That was the predominant mindset and, although it was much better than not looking for causes at all, it was inferior to problematic thinking - or tracking the clues that led to the causes.
Underneath this idea of problematic thinking lie several important Six Sigma principles, including the precept that the output (Y) is a function of the input (X). Beyond this, we know that to improve Y we must improve X, we must measure both Y and X and we must understand the causative function (f). By focusing on the critical product characteristics and related process capabilities in this manner, a corporation can solve the problem of reducing cost and providing higher value to its customers and itself, simultaneously, in an a priori rather a posteriori fashion.
Perhaps the best way to get at the heart of the difference between symptomatic problem
solving and problematic problem solving is with an analogy. Dr. Harry is a volunteer member of Sheriff Joe Arpaio's mounted posse in Maricopa County, Arizona. Known as " America's toughest Sheriff," Arpaio takes a punitive rather than rehabilitative view of crime. The prisoners in his jails work on chain gangs, are required to wear pink underwear, are fed two rather than three meals a day and are stripped of many other privileges enjoyed by most other criminals in America's penal system.
If you ask Arpaio why he goes to such lengths in humiliating his inmates, he'll tell you his primary objective is to make life tougher on the inside than it is on the outside. Stands to reason that, maybe just maybe, a potential repeat offender would think twice before committing another crime that might land him back in a Maricopa County jail - or in Arpaio's famous "tent city," where convicts are exposed to the brutal Phoenix summer heat.
Dr. Harry's natural inclinations as a cowboy led him to Arpaio's posse, a group of Deputy Sheriffs that conducts Search and Rescue activities. A large part of the Sheriffs posse fights crime on the streets and in police cars, but Dr. Harry signed up for working the hard-to-access areas of the Arizona desert the old-fashioned way - on horseback. Wouldn't you know it, his job in the mounted posse largely involves the skills of a tracker - a person who looks for clues that reflect the condition and whereabouts of the person who is missing. In a proper search, the tracker doesn't look for a person; he looks for the tell-tale signs of the person, where he has been, the direction he is heading, what he might be thinking and what he is doing to get out of his circumstance. For example, a person who is throwing chewing gum wrappers along his path is probably not in imminent danger. The best trackers follow a carefully prescribed "pattern of search" that fully considers the topology of the terrain - much like a Black Belt follows the "data path" of a problem.
As a tracker, paradoxically, if you look for the person you will miss the clues that can lead you to where he or she is. It's the same way with solving problems. If you look at the symptom, you will miss the clues that can lead you to the problematic cause. When we talk about clues in a corporation, we're talking about performance measurement data, and about giving proper due diligence to the study and analysis of that data. We are torturing it until it tells us what it knows and points us to where the true root cause of a problem is likely to be hiding.
While training at Motorola later in his career, Dr. Harry used to show a transparency slide that read, "Let the data do the talking." The visual part of the slide showed a magnifying glass looking at footprints - looking for clues. With this slide, he was promoting the idea of spending 85 percent energy on letting the "data-based clues" frame the proper questions by which to further interrogate the clues, connecting one to another until the solution became self evident. The other 15 percent energy could be focused on recognizing, defining and measuring the symptom.
Dr. Harry remembers one very bright woman who worked on his production line at a GM factory. She was so knowledgeable about the different kinds of defects that she would keep a mental Pareto chart in her head. Sometimes she took quality so personally that she would shut down the work at her station when there was a problem and come to Dr. Harry with her thoughts about what to do. Apparently, she expected him to do something about it.
What she didn't realize is that the system was not reinforcing quality intervention or defect prevention. Naturally, Dr. Harry tried to influence local management by pointing out these quality problems and advocating a disciplined approach for preventing them. But his admonitions fell on deaf ears, as the existing system was predicated on the principle that it was cheaper to scrap bad product and keep the line running.
For example, in Dr. Harry's area, the plant would build up pans full of bad field coils for repair during schedule lulls. You see, there was a union agreement that people were sent home with pay during schedule lulls or unanticipated shutdowns. But if you had a slew of repair work sitting on the sideline, then you could keep people busy during these times rather than send them home to get paid on what was referred to as "F" time - factory time. In a twisted logic sort of way, bad quality was good because it gave workers something to do and, therefore, constituted a form of productivity optimization. After all, if the factory made every product perfectly, then what would the workers do during down times? The company would lose money because people would be paid for doing nothing while sitting at home.
The only word Dr. Harry could think of was "unbelievable." Didn't management understand that if the line could produce high quality consistently over time, then fewer overall workers would be needed? Any downtime labor hours paid in this scenario would be far offset by the overall payroll savings. The forest truly was lost in the trees, and detrimentally shortsighted thinking dominated the day.
Dr. Harry felt sorry for the woman, the one and only person in his department who took quality personally. He would placate her and deflect her legitimate concerns, as was his job, although inside he didn't feel right about it. Turns out that same woman reached a point in life when she could afford to buy her first brand new car. It was a Pontiac, and it wouldn't start the very next morning after she bought it. The problem was a shorted field coil in the starter motor, originating in the very department where she worked! The person who cared the most about quality got burned in a rather ironic way.
Peace at any price. Interestingly, it wasn't just manufacturing that allowed defects to pile up "for the good of the company." Even in labor relations, grievances would be allowed to pile up in the spirit that there is power in numbers. In other words, unresolved grievances are good because they give you more leverage when it comes time to renegotiate the labor contract. As in manufacturing, the emphasis was on quantity, not quality. Resolving grievances as they occurred was discouraged right along with fixing defects in line.
Much as manufacturing delivered its quotas at any price, in labor relations it was a time of "peace at any price." The union shop stewards would collect grievances on the basis of hours until the point at which it made sense to sit down with management to settle them, such as during contract negotiation time. Together they would work through each grievance assigning it a certain number of labor hours until they came up with an aggregate number of hours in the end. By consent, management empowered the stewards to distribute the bulk of hours (which translated into dollars) in whatever way they saw fit. For example, management might settle 100 grievances with 1,000 hours of pay. In turn, the shop steward would go back and distribute the hours to his people, and sometimes to himself, as he best saw fit.
So, in reality, the steward might have gotten 30 hours for Mary's claim, but he would only give her 25 and keep five for himself. Naturally, these hours represented the "settlement" and could be turned in for cash. Perhaps that's why so many shop stewards could afford to drive new Cadillac's — because it was peace at any price, and that included the practice of grouping grievances and "batch processing" the settlements. This kept the machine well oiled and running. There was a process, but it wasn't a quality process. It was a simple-minded way of ensuring the corporate value of meeting schedule (at any cost) and avoiding the interruption of labor disputes and strikes.
After having resolved their grievances in this fashion, the various operational divisions would carry the following message to corporate meetings: "See how good we are, we don't have any grievances. We are keeping Pontiac running, Buick running, Cadillac running. We are meeting our schedules, we don't have any unhappy people and we are able to get along with the union." Such a message, backed up by "the numbers," was good enough to earn division and plant managers a nice bonus. It was peace at any price, and they had long figured out how to buy such peace, and false prosperity.
Indirectly, plant managers were buying their schedules by operating the hidden factory. They knew, for example, that inventory was measured in May and that they had F time in April. So rather than fix quality problems on an ongoing basis, they would let them pile up for the time in April when they knew they would be down. Then, rather than sending people home during that time, they would fix the defects in April and get it back into production to meet the schedule. This way, when May rolled around, there was no excess inventory. Everyone looked good because there was a means for covering up poor quality.
As a consequence, the only visible issue was scrap, and often this too was covered up. For example, if a plant manager had a bunch of field coils that were mis-wound (defective), he would cut a deal with the QC manager, who would say the cause of the problem was due to bad copper wire. This way, the cost of scrap became chargeable to someone upstream in the process, like a supplier, not the given production department. In return for the QC manager's favor, the plant manager would cooperate at times when poor quality could potentially make the QC manager look bad.
It was a classic case of you scratch my back, I'll scratch yours. The plant manager wins and the quality manager wins. The only problem was that the customer, and the company, would lose because the overall operation was sub-optimized. It was a climate of looking good and feeling good, regardless of reality. It was a climate of hear no evil, see know evil and, therefore, evil doesn't exist. It was the corporate equivalent of the "I'm OK, your OK" era.
Believe it or not — cloaked in "legitimate" rules, regulations, operating procedures, goals, objectives and requirements - the same operations management scam is going on in many corporations today. Just as they did back then, so many of today's managers know how to "work the system" to make their areas of responsibility and themselves look good. They know how to cover up or sidestep poor quality, particularly when the key metrics by which they are judged do not correlate to the key parameters and tenets of quality.
In the old days at GM, when Dr. Harry ran a field coil department, he was not held accountable to any particular quality metric. He was symptomatically measured by his scrap rate, and he knew how to play the game and work the system. He turned his red tags (batch of production tagged by quality department as bad) into green tags by cutting deals with the QC manager. Compromised product was then shipped to the customer because, after all, it was under warrantee. Feedback from the field was so sluggish that it wasn't uncommon to hear about failures a year or more after shipment, by which time the typical "good" plant manager had already been promoted out of the area.
This is how a politically astute plant manager, or department manager, took advantage of passing poor quality on to the customer: by cloaking the costs of poor quality in a never-ending stream of price adjustments. Again, it was a game of utilizing system weakness to optimize one's perceived performance as a manager, while at the same time sub-optimizing the exchange of value between the customer and the provider. It was a scenario in which value entitlement was not being realized for the customer and the provider in every critical aspect of the business relationship.
As foreign and far-removed as this game may seem, it is still played by many managers in many corporations today. The exact metrics and loopholes aren't always the same, nor are the names of the players, but the principle of sub-optimizing the whole to promote the few is still in place. If you don't believe it, then ask yourself how certain corporations continue to reward their managers while their cost of quality and defects remain relatively constant over the years.
Problem-Solvers as Hired Guns
Toward the end of his four-year tour at GM, Dr. Harry ran across a yearbook from the Council of Technology Education. The yearbook was a compilation of papers written by different professors on a set theme and read by educators and technologists at the university level. It was one paper in particular that piqued Dr. Harry's interest and inspired him to set his life on a new direction. To this day, Dr. Harry wishes he would have saved that paper so he could quote its source, but at the time he didn't understand the significance it would play in his journey to Six Sigma.
Here's basically what the paper said: with the explosion of technology in the world, there is also an explosion of problems. In other words, the amount of technology we create fosters complexity which, in turn, contributes to the creation of problems. The essence of this is the adage "to answer a question generates three more questions."
As the world becomes more technology-rich and complex in terms of products and processes, universities will not be able to sufficiently teach all that is required to master a profession and to be effective problem solvers. Yet engineers and technologists are expected to solve complex problems soon after they enter the workforce. Therefore, the paper asserted that the future elitist jobs would be filled by those who are master problem solvers - technically oriented people who can cut through the ambiguity of complexity and make it their slave.
The paper went on to say that the principal skills of such people would be mathematical statistics overlaid on a generalized technology background and a deep working knowledge of the scientific method, as well as key business practices. In this sense, the most valuable employees of tomorrow would be fundamental jacks of all trades, masters of none, but expert problem solvers who could also teach others in a geometrically expanding way. The best of the best would be those who could wield the power of data for the benefit of their corporations and themselves, and propagate that benefit to others in the form of mass training. They are the ones corporations would recruit, promote and provide the opportunity for very rewarding careers. Indirectly, the author of this paper was describing a hybrid role - much like that of a Champion or Master Black Belt (as we know it today).
In addition to leveraging data and statistics to solve problems, the paper also said that the future lies in the hands of those who can teach others what they know, and provide the necessary leadership for ensuring the application of that knowledge. Recognizing that people can't do what they don't know, knowledge becomes a tool of empowerment. Therefore, if you own problem-solving knowledge, you can train others to the point at which it has a multiplying effect throughout a corporation. Having been a commissioned officer in the Marines, Dr. Harry understood all too well the immense value of such people - leaders who are willing to share, not stockpile, their expertise.
To make a long story short, Dr. Harry was so deeply stirred by what he read in this forward-looking paper that he decided to leave GM and pursue graduate work in, guess what? You got it - technology, problem solving and education. Driven by ambition, Dr. Harry recognized that people came and went in corporations, but the ones who had staying power were the ones who could successfully and consistently solve problems and then pass that knowledge on to others. Regardless of the mistakes they might make in their respective disciplines and areas of responsibility, if they consistently solved tough problems they were almost always protected from the chopping block when it came time for layoffs, and at the front of the line for promotion.
The only problem for Dr. Harry was that there were, per se, no graduate programs in problem solving. The need was simply too new and on the edge. So he went and completed a Master's degree in technology and education with independent study in statistics and problem solving. Then he realized he would have to design his own Ph.D. curriculum if he wanted to acquire the academic credentials for becoming an expert problem solver. For this he turned to Arizona State University (ASU) in Tempe, Arizona - specifically to the school of engineering and applied science, technology division.
His self-designed but committee-directed curriculum consisted of courses in management, education, instructional design, technology, psychology, statistics and science. Also during his doctoral work, Dr. Harry made a point to develop and test a problem-solving methodology that he could brand to his own name, then plant into the corporate world. In doing so, he would secure a very special niche for himself as someone who could do what others were not educated or prepared to do.
His educational objectives were several-fold, including the mastery of problem-solving systems, statistical data analysis, experimental design, knowledge design and transfer, as well as management and psychology. His educational strategy was also several-fold, including coursework, independent study and on-the-job application. These were the means by which Dr. Harry would achieve his objectives, but only with the kind assistance and patience of his doctoral committee, which patiently put up with his constant need to break the mold of the familiar.