Benefits of Process
The word "quality" means different things to different people. For some, it's a sign of endurance. For others, it's a measure of innovation. But just about everyone agrees that, in the end, the concept of quality has to do with expectations, usually the expectations of the customer.
When you build something that performs in a way the customer wants or needs, you'll typically be credited with delivering quality. If you miss the markno matter how well put-together the thing isits quality quotient will come into question. In the business of technology, then, it's important that organizations arrive at a definition of quality. The value in this bit of advice becomes clear when you think about process.
Process is about inputs and outputs. It is a way to shape inputs in order to generate desired outputs. In any production system, the ultimate output is a viable product. "Quality" then will be imbedded in the thing you've created. And so any process is dependent on first understanding what the output has to be.
The phrase "quality culture" describes an organization that has designed the pattern of its activities to deliver a product (or service) that carries within it the corporate definition of quality. Very few IT organizations would admit to having a weak understanding of what their customers want. And even fewer would admit to having no real way of getting there. But more than a few studies indicate that definitions of quality and the production processes that help ensure quality are absent in many, many IT shops.
Many American technology companies have been slow to embrace quality cultures. I think I know the reasons why; it's a blend of issues. When I have sat with CIOs and other senior IT managers, I hear pretty much a common theme. They listen politely to me and then say, "That's great, James. We really see the value in that. But right now we've got other priorities to wrangle with. This just isn't a good time for us." What can you say to that? If management is focused on pressing issues outside this realm, there's probably little that will change that focus somewhere else. The best follow-up may be to schedule another visit.
Another reason for the hesitance is that U.S. technology shops have traditionally been held pretty much unaccountable for the specifics of their operations. At the macro level, they need to hit a certain operating budget, and they need to solve general problems, but what goes on inside the black box is not readily scrutinized. I touched on this in the introduction. External managers still think in large part that technology is a mystery, that it's complex engineering, that it's a specialty realm best judged by those on the inside. As long as what comes out of the box works, it almost doesn't matter what it took inside the box to make it happen. In these environments, building it right the first time and fixing it 12 times prior to release are the same things.
And then there's the simple question of education. To my surprise, I've discovered that many of the managers running the $350 billion technology industry in America have almost no knowledge of quality management. When they think of terms like "quality assurance" or "quality control," they automatically think testing. They know a few of the rules"you can't test quality in"but they don't move much farther from there. It may be that the renewed push to quality is a recent thing (what with Sarbanes-Oxley and all). And it may be that few if any of these managers ever received any training in the fundamentals of quality management. By and large, IT managers move up in their organizations through a technical or engineering path. They just haven't had exposure to this body of knowledge. Those that have acquired it typically do so through their own initiative.
What I am convinced of is that once people understand the benefits of implementing a quality program, the reasons not to do so pretty much evaporate. And there are plenty of reasons. Some as we have noted are esoteric and qualitative, but many are tangible and concrete, and translate directly into improved performance.
The benefits of process management, process focus, and process improvement are real. They are not abstractions or reductions. They are not derivatives of some other result. Process works. And not because it is a revolutionary innovation. The opposite is true. It works because time has shown that all production is a process. If an organization employs weak or misdirected processes, its products will be produced in a weak, unpredictable fashion. If an organization employs sound processes, its production environments will operate in smoother, more controlled fashions. In that way it is rocket science. Rocket scientists use some of the most efficient and proven processes ever put to work.
We looked earlier at some of the more empirical, quantitative benefits of processsuch as ROI numbers from Boeing and Schlumberger.
But here are some just-as-real benefits of process, of a more qualitative nature but providing the same significance of impact.
I was in Manhattan not too long ago visiting a friend. Shelly's a financial analyst for one of the larger brokerage houses. One evening she took me uptown to a restaurant off East 52nd street to meet some of her work associates. I expected an evening filled with talk of P/E ratios, leverage points, and acquisition targets. But I was surprised at the interest Shelly's friends had in what I do. They seemed to know quite a bit about it. They were quite familiar with how Jack Welch used Six Sigma to get a real handle on defect reduction at GE. And they knew how Motorola has used it to boost pager production. They talked about CMMI at Lockheed Martin and Northrop Grumman, and the government's growing insistence on the use of quality systems by its major vendors.
"When did you guys get into process?" I asked Shelly.
She tapped her finger on the solid surface of the table. "James," she said, "stability is an asset."
That was the first time I had heard it put that way. But Shelly was right. And her analyst friends had picked up on an intuitive insight. They were beginning to add process into the mix of elements they used to value companies. It made perfect sense. Stability certainly is a corporate asset. And process is a way to add stability into an organization. A process program can help set boundaries for operations and for production. It can establish paths of activities that your work teams can follow in a consistent and repeatable manner. It introduces predictability into work flows. It helps you forecast and plan. And it does this in a way that is visible to everyone in the organization, from management on down. Further, it provides a mechanism for the public to see this stability as well. And that can go a long way to enhancing the image of an organization.
Over the years, it had grown its software development services in fits and starts. Between the ups, it focused on staff augmentation, farming contractors out to other companies for a cut of the hourly rate. So it did some contract development and some contract placement. Then, because it did have some pretty senior technical folks in-house, it began to find revenues in doing architectural assessments for other software development companies, sometimes its competitors. Then it found some luck putting together and farming out Test Teams. After a dozen years of business, Tenzer found itself in about half a dozen lines of business.
The problem it faced was that it had no real identity. Tenzer was whatever it was that could bring Tenzer revenue. Naturally there was very little staff cohesion, little strategic focus, and almost no loyalty. Tenzer turned into a marriage of convenience among 100 or so partners.
I don't know if some form of process early on could have helped Tenzer. But the case illustrates well the importance of identity. Process is one way to help an organization establish its identity.
You can see thisalmost without lookingat very successful process companies. There is an indisputable way of doing business at Coca-Cola. It is the Coke way. The same holds true at Apple Computer. Apple's way of doing business was not born out of its marketing image; its image was born out of its business processes. Think of Kentucky Fried Chicken. Or NASA. When you embrace a process program, you embrace a set of practices that evolve over time into a set of cultural behaviors and values. Those things that your processes emphasize and promote become the culture.
Think of the phrase, "There's the right way, the wrong way, and the Army way." That's a process-driven mantra. Corporate identity may be just as important an asset as operational stability. Identity is what keeps a collection of individuals (the basis of any company) together. It is what helps move them forward in a common direction. Process can make a significant contribution to establishing this identity. It takes what can be an invisible, intangible attribute and, through action, makes it visible throughout the whole organization.
Quality management programs save money several ways: by reducing costs, by streamlining operations, by helping shrink waste. But these actions not only save money; they sharpen the focus of an organization.
Every organizationevery functioning grouphas a culture, and that culture dictates how the company works. Some cultures are open and free-moving. Others are structured and carefully balanced. But either way, the culture is impacted by how the company is focused. You'll find that a quality management system is an effective tool for focusing the energies, work direction, and activities of your company. Let's look at an example.
Burger King has a quality management system. That's why a Whopper ordered in Tucson tastes exactly like a Whopper ordered in Jacksonville. Thirty years ago, Burger King figured out how to make Whoppers, and by and large, they've stuck to that successful formula. Burger King's employees and managers don't have to figure it out anymore. The inventing part is done. Now they are free to focus their energies on other things: new sandwiches, new promotions, the future. The burger-making issueaside from occasional refinementshas been taken care of. The formula is established.
Operational unity is a desirable trait because it sharpens focus. An organization has only so much energy it can expend to meet its goals. If the energy is dispersed, like in a soft light bulb, the light can only travel so far before it dissipates. If the energy is concentrated, like in a laser beam, the distance greatly increases.
Through an effective quality management program, you can sharpen the focus of your organization. Your people don't have to continually figure out effective ways of doing businessyou've already found those and documented them in processes. Instead, your people are more free to innovate, to move the company forward toward its goals.
The organizational unity that comes from a sharper focus has another advantage: it solidifies the organizational culture.
If you look at hyper-successful companies like Mars, Johnson & Johnson, or Apple Computer, you'll find that they each have a very distinct culture. Were you to work there, you would know that you were working there. There's a feeling about each place, and that feeling comes from their way of dong business.
Mars, Johnson & Johnson, and Apple Computer have worked hard to establish cultures that reflect their values and goals and their ways of doing business. When you think about it, a quality management system helps you do the same thing. It formalizesexternalizes through its processesyour goals. As a consequence, when you're working in an organization with a defined process program, you'll find that it's easier to know what is expected of you, to know what contributions you are there to make, and to know how to work effectively with other people and other groups.
There is a popular management exercise called JET, Job Expectation Technique. The idea behind JET is that members of a group can't work effectively as a team until they understand four fundamental things: the mission of the group, their job within the group, what contributions they are expected to make toward the mission, and how to communicate with other members of the group. Once a person understands his place in the groupand the value he brings to the grouphe can become an active participant in that group.
A Higher Performance Bar
In any organization, you get a mix of talent. There are strong performers. There are weak performers. The committed and the coasting. Thinkers and tinkers. That's the mix of human nature. Management's job is to bring together the different people that make up an organization and harness their different degrees of talent toward a common purpose. To a degree, that mission exists with or without a quality management system or a process program. But when looked at from a best-practices viewpoint, a quality management system can play a complementary role in that missionand not a passive role, an active one.
By definition, a quality management system is a collection of processes selected for suitability and designed for effectiveness. The processes are best when they are built (at least in part) on the best practices of your strongest people. The chief quality, then, of a proven processone proven to facilitate a desired resultis that, if you follow it, it should deliver predictable results. In other words, if your best employee follows it, it should produce pretty acceptable results. And if your weakest employee follows it, your chances are strong that you'll at least get something similar. Organizations that create effective process programs almost always look to their strongest performers for input, guidance, and feedback. The program then ultimately reflects how those guys, in large part, work.
When you set a quality management system into place this way, the program begins to lift the performance bar for the whole organization. It gives workers discrete accountability standards for their own work. At the same time, the internal success secrets possessed by the talented and the committed become externalized and available to all. The process program you build as a tool to manage quality becomes in use an organizational educational tool. Its themeonce you have sharpened it through useis "This is the way that our organization works." Your program begins by setting a work path for all your people; it allows them a predictable route on which to practice their talents. Further, it provides the resources and training for all team players to become effective contributors. Redundant explorations and repetitive seek-and-find are greatly reduced.
Through a quality management program, you'll find that your super-talented people are free to spend their time on innovation, on creativity, on pushing the company forward. Dan Payne, a senior IT project manager in the telecommunications industry, specializes in helping IT organizations use process to free their best folks to focus on innovation. Dan works chiefly with CMMI, and in these implementations, he reports that, in addition to freeing the best, you can take once-mediocre performers and give them the structure and the guides they may need to become valuable players. At this point, the core talent of the organization is no longer embedded invisibly in its people. Your quality management system exists, in large part, as a success program everyone can follow. When you design your program to reflect successful practices, when you school your people to follow these practices, and when you conscientiously commit the organization to the path, you provide a real service to your people. You raise the bar for everybody. And you help them clear it. Performance rises.
The process prophet Philip Crosby coined the phrase, "Quality is Free." If there is one misconception about quality that I encounter more than any other, it's that quality costs. To many, implementing a quality program means adding overhead to an organization. It's seen as an expense with a questionable return. Maybe you know people who think this way. I don't mind this opinion too much because it is easy to overcome.
Here's one illustration:
Around 1980, Art Sundry and Bill Smith were in engineering at Motorola, making pagers for the company. They were working on the issue of defect management because the company had been struggling with a quality problem for some time. They were looking at the pager production line to see if they could learn anything about how to make things better. They followed pager assembly down its many steps. Once a pager was put together, it went through a pretty comprehensive final field-ready check. If it passed the check, it was packaged and shipped. If the test hit an error, the error was noted and the pager was sent to repair. Once the pager was repaired, it was repackaged to be shipped.
Sundry and Smith began to keep their eyes on these early trouble pagers. What they found was these pagers carried their troubles with them. Pagers that were flagged to go through the repair process turned out to have significantly more problems in the field than pagers that passed. Put aside the issue of testing thoroughness or flagging procedures right now. The point is that the products that left the shop in a quality state retained their quality state in the field, well into life. The products that showed early signs of problems kept on evidencing problems, even when the original ones were corrected. Fundamentally, quality products required less attention than defective products.
In fact, many additional production steps were required to ship a product flagged as defective:
And that's just five basic steps. The procedure at Motorola was no doubt more complex.
What was this repair route costing Motorola? In labor, in inventory, in time, in market reputation? There were all kinds of possible ways to address this problem. Open more field repair centers to reduce customer downtime. Subcontract out for ready-assembled components. Farm all defect repairs out to a centralized operating group.
But Sundry and Smith, early pioneers of Six Sigma, knew that the best solutionthe right solution for Motorolawas to get to the source of the issue, to reduce defects, to produce pagers that passed their initial quality test and thus promised reliable performance in the field. The way to cut costs and save money at Motorola while strengthening its market position was to raise quality.
The complexities of moving phone numbers from, say, AT&T over to BellSouth were enormous and involved questions of reprovisioning, new switch addresses, node transfers, and on and on. And so we contracted a company out of Ottawa, Systems House Limited (SHL), to code the system as we designed it. SHL had a great deal of telephony experience and was very familiar with the demands of such number-sensitive operations. What we didn't anticipate was SHL's meticulous approach to managing this project.
SHL was an ISO 9002 registered shop (a direct antecedent to ISO 9001:2000), and they had every intention of following their quality management system for the life of this contract. We balked at first (and I was one of the balkers). SHL wanted to do a lot of planning. They wanted to spend a lot of time validating the design materials we gave them. They asked a lot of questions and got their teams (and us) together often to drive down to detail. They had a great reputation as a coding shop, but it often looked like the last thing they wanted to do was code.
We worried because Local Number Portability was a federally mandated service. Telecommunications companies had to offer this free market service to its customers by a certain date or face major fines. Between now and the due date, we had to make a lot of smaller deadlines: industry review meetings, the exchange of coordination plans, pilot runs of data transactions, etc. We didn't want a lot of additional "management activity" that might slow us down and potentially make MCI look less than competent to its rivals. But SHL insisted, and we went along, more or less because by that point we had no choice.
But then we began to notice something. The cross-industry design meetings began going very smoothly. Our teams seemed to have a deeper grasp on the issues and a clearer vision of how things should flow through the system. Our designs became stronger, and the picture of the whole system with all its interrelationships began to take solid shape. When SHL began to code, the components worked just about how we had all anticipated. There were some backtracks and some redos, but by and large the construction phase was remarkably uneventful.
Two months later, the government pushed the deadline back because most of the other phone companies weren't making our kind of progress. That just gave us more time to fine-tune our solution. When it came time for integration testing, the system proved to be in solid shape. When we did system testing, it was even tighter. When it came time for acceptance testing and production dress rehearsal, MCI was able to forgo bringing in a large support staff. It had planned to hire upward of 70 contractors for a final blitz, but there was no need to blitz. A team of about one-half that size proved adequate.
A little over two years later, MCI WorldCom would go down in a blaze of flames, but in that project MCI led the industry in readiness, innovation, response, and quality. And there's no way to estimate how much they saved in hard costs, aggravation, and industry stature by working with a company like SHL, a company that insisted on doing things the way they knew was right.
The Project Is the Process
This may be one of the strongest benefits that process can deliver into a technology development organization. The full breadth of this benefit didn't really hit me until I had been in the business for about six years. But what I began to see was this: IT shops that had used process for quite a while seemed to be able to very confidently and consistently predict budgets, schedules, activities, contingencies, and variances. They seemed to have a good grip on things that at first, to me, appeared transparent. I was impressed that these places could visualize a project, with its different dimensions and demands, very early on in the formulation period. But after a while I saw the trick. Only it wasn't a trick. It was a technique. And a pretty reliable technique.
The people in these organizations had begun to place the concept of The Project behind that of The Process. In these shops, process superseded the project. When I looked at their processes, it was clear why they were taking this approach. Their processes had been well tested over time, so much so that these people could rely on them and depend on them. And when you looked closer, you could see that they had a process for every major phase in technology development. There were processes for requirements management, requirements development, technical design, peer reviews, defect handling, verification, product integration, validation, and delivery.
When these shops acquired new project work, they began by assembling all the processes they would need to get the work done. Once the right process set was in place, the full set told the project story. For all processes, there were inputs and outputs. These were planned as work products. For all processes, there were steps. These were planned as activities. Resources were assigned to support these activities. Timelines were set into place. Budgets were allocated to cover the costs.
By the time the process assembly was completed, the project itself had taken full shape. With a process program in place, your organization can count on the same kind of advanced step-ahead. Your way of doing business becomes encapsulated in your processes, and so you can more readily predict the shape of your projects by the processes you put to work on them.
Sure, why not. Seems obvious. But it's worth setting out explicitly because while many of the benefits of a quality program are by-products of the program, better quality is the primary product. And better quality has such a radiating effect that it can almost be said to be its own reward.
We discussed the question of what quality is earlier in this section. Once you define what it is and how it should be realized in what you produce, and then work to realize that, your company almost automatically becomes stronger.
Quality enhances your company's reputation. Quality has a tendency to motivate the work force, inspire innovation and creativity, and bring out best efforts. Quality leads in the marketplace. It promotes sales, builds customer satisfaction, and fosters brand loyalty. It establishes competitive standards. In many ways, for many industries, it defines success. It's the benchmark for achievement.
The old adage "nothing succeeds like success" is true. And it's a derivative of delivering quality.
The art of process improvement revolves around three concepts: establishing a process program, using the program, and evolving the program's effectiveness over time. Each of these requires its own approach and its own set of considerations. The firstplanning the scope of the programappears to most people to be the most daunting. In the following chapter, I'll show some of the factors you might consider when setting out to develop an internal process improvement program.