Six Sigma for Software – A Role Based System

Because the application of the Six Sigma methodology is relatively new to Software and IT, many of the readers of this web site may be unclear as to the methods of deploying Six Sigma into an organization successfully. From participation in the discussions section at iSixSigma Software, it is clear that some confusion exists about the effective use and positioning of Six Sigma in a Software organization or function. This article is intended to help the reader better understand the early attributes of how Six Sigma is typically and successfully deployed inside a Software organization. It will be part of a continuing series exploring concepts and options of Six Sigma deployment designs and best practices in Software Organizations.

The Investigation and Commitment Phase

In many cases interest in Six Sigma starts as a minor point of curiosity. A few individuals either read an article or attend a conference and become interested in the significance of the results or the importance of the Company discussing those results.

Success in Six Sigma is almost always expressed in financial terms. Because of the recent economic conditions in the software industry, any significant cost saving metric is almost certain to cause an immediate and “eye-catching” phenomenon amongst most managers and executives. Frustrated by the lack of progress and efficiency, despite significant spending on tools, training and processes, management is always on the hunt for a new fix or an answer to their concerns. This in itself is part of the problem. What is needed, and what Six Sigma brings, is a systematic approach to prevention not another isolated ad-hoc fix.

Software is still lagging in terms of efficiency, quality, cost and customer satisfaction. However, uneducated decisions by these very managers may have been the cause the problem in software to begin with. Another uneducated decision about how to implement Six Sigma will only compound the problem. In response to this we need to be diligent in how we educate interested managers and executives relative to the truths and myths of Six Sigma.

As many of the readers now know Six Sigma is NOT a quick fix, a single tool, purely statistical or a new fad. It has been around for 15 years and has evolved as an integrated part of a practicing companies management system. It does not replace other initiatives or technology. We do not apply Six Sigma to everything. It is however a powerful methodology to bring a rational, common sense approach to improving cost, efficiency, quality and cycle time, based on analyzing and improving the processes that produce the work product. To maximize the long-term impact of a Six Sigma deployment on an organizations performance, the decision makers must learn early and often that their role in Six Sigma implementation must be active as opposed to passive. They are the key to selecting the resources that will lead Six Sigma projects and must drive the process of aligning and prioritizing when, where and how Six Sigma projects will be implemented.

For these reasons, successful Six Sigma deployment plans usually start with a series of Executive Sessions to orient the Executive team and plan the deployment. This is extremely important in Software because there are many traditional practices in software project management, which compromise the proper investment in up-front planning without considering downstream consequences. This is not done intentionally. It is simply the state of an industry trying to speed up progress. Sometimes we need to slow down to speed up. One example of this is software project estimation and resource loading. This problem is usually characterized by the lack of a process to understand accepted resource loading models for varying sizes of software projects (which are based on widely available and accurate data). This generally leads to poor estimation and improper resource loading creating downstream software bugs and the added (and unbudgeted) costs to fix them. Through the combination of specific estimating models and Six Sigma statistical tools we are able to make better decisions about resource loading, improve the accuracy of estimations, reduce overall cycle time and reduce lifecycle cost. (There have been several good articles and discussions on this web site that speak to the details of this topic.) The point of all this is that these are all management issues. Until management understands and acknowledges these issues it is very unlikely they will be addressed. Six Sigma finally provides a proven roadmap and methodology to drive a change in this system. And a key to that system is the specific roles, responsibilities and accountabilities of the Executives, the Management Team and the individual contributors.

Six Sigma – A Role Based System

Six Sigma is many things to many people. But one of the key attributes that help to make it successful in today’s stretched organizations is its role-based nature. In Six Sigma we have defined clear roles for various parts and layers of the organization. These roles have provocative names including Executive Champion, Deployment Champion, Champion, Black Belt, Green Belt, Yellow Belt and so on. These are more than fancy names to get attention. They are specific roles for the various layers of the organization with defined responsibilities, time allocations and accountability. Without the accountability and full engagement of each person, the system will not function as intended. This is not unique to Six Sigma. It is true of any major organizational initiative. Therefore, organizations that have historically had successful deployments of other large initiatives may be able to apply lessons learned to the implementation of Six Sigma. These characteristics would include discipline around items such as Communications, Project Metrics, Resource Allocation, Budgeting, Project Review, Reward and Recognition.

Figure 1: Six Sigma Organizational Layers

Champion Role

Six Sigma organizational roles related to typical organizational layers are depicted in figure 1 above. At the top of the organizational pyramid are the Senior Executives and / or top managers. They usually assume a champion role of some kind. Generally champions do not spend additional incremental time on Six Sigma (other than some up front training and planning). They instead, learn how to reallocate current time from reactive problem solving to aligning critical business issues to metrics and then achieving permanent solutions to those issues by assigning and supporting the right resource (Black Belts/Green Belts), equipped with the right tools (Six Sigma Methodology) and giving them the time and support to accomplish that objective. While this may sound like common sense, research data indicates that most Managers spend between 50% and 70% of their time reacting to problems. So much so that, for many, this is simply viewed as the job at hand. In fact Managers in “best in class” organizations spend more time planning, prioritizing, aligning resources and defining problems than their counterparts in highly reactive organizations.

In a previous iSixSigma article Six Sigma…More Than a New Tool we discussed the need for a more systematic approach to improvement in Software as opposed to the current ad-hoc approaches and “programs of the month” which have not yielded the advertised results. The deployment of Six Sigma is very much a “back to basics” approach. And, it starts with Management. Without an active management role, with consequences attached, long-term behavior will never change. In practical terms this means that Six Sigma Champions must initiate a process to keep the Six Sigma system filled with projects and people. It is the Champions role to define, prioritize and scope Six Sigma projects and determine what resources will be assigned to them. It is also their role to keep those project clusters aligned with critical business issues. As the system ramps up, the Champion must maintain an ongoing role to review project status, results and issues as well as removing barriers to problem solution implementation. If this is done properly and pervasively it has been demonstrated to break the chain of firefighting behavior and replace it with data based, prioritized and preventative culture.

The Black Belt Role

The second Layer of a Six Sigma deployment usually involves middle management and the organizations professionals. Examples of job titles considered for Black Belt Positions in a software environment include (but are not limited to) Software Engineer, Product/Process Improvement Specialist, Project / Program Managers and Software Quality Specialists/Engineers. Because Black Belts are a full time resource an organization must carefully consider and have a system for, choosing, compensating and creating a succession plan for Black Belts. These should be considered prestigious assignments reserved for an organization’s fast track high potential employees. As a rule of thumb, if a person is “available” they are probably not the right person for the job. Why, because this position is going to be responsible for solving the organizations most important and challenging problems. The people you want for this job are probably some of the busiest and most important people in your organization. But, because they lack tools, priorities, and time they are spending most of their time firefighting. This creates a high level of frustration and lack of a sense of accomplishment.

This high-pressure environment contributes to another industry issue, lack of employee loyalty and high employee turnover rates. The decision to commit a person to a Black Belt role on a full time basis is probably the most difficult role commitment of all in adopting Six Sigma. But at some point, if an organization wants to break the chain of reactive behavior, disappointing results and poor quality (leading to excessive costs) it must do something dramatic. That is why, to many, Six Sigma is known as the “Breakthrough Methodology”. It facilitates a breakthrough in the management system of our people, processes and results. These individuals will be assigned high priority projects to complete during their training. Completion of these projects generally yield a cost savings and/or avoidance between US$100, 000 – US$1,000,000. Typical Black Belts will complete between four (4) and six (6) Six Sigma projects per year after completion of their training. It is typically a 2-year assignment at which time they should be moved into the next logical job on their career paths. In most Six Sigma organizations a range of about 0.5% -2% of total employee headcount are Black Belts, depending on their level of commitment, business needs (cost savings targets) and size of the organization.

The Green Belt/Yellow Belt Role

Six Sigma is not simply a group of elite individuals solving all of the organizations problems. Black Belts do lead and facilitate teams through the use of the methodology and ultimately will get the input of a team (usually comprised of product and process owners and individual contributors) in the definition, prioritization and implementation of problem solutions. What we have established as reactive behavior extends throughout all organizational layers. So, what is the role of those who are not full time Black Belts? Two roles have been established in the Six Sigma system. Green Belts and Yellow Belts.

Green Belts are in many ways similar to Black Belts with one important difference. They are not a full time Six Sigma resource. In Software, Green Belts usually hold positions such as Software Project Managers, Software Designers, IT Support Positions, Software Engineering Positions, etc. These are people who by the nature of their jobs must solve problems on a daily basis. However, prior to Six Sigma implementation their problem solving is likely done in an ad-hoc manner with a limited tool set. Their training in the Six Sigma Methodology gives them a practical and proven method and set of tools to apply to their daily problem solving activity. They may require help or mentoring from a Black Belt but usually, after completion of training, they apply the Six Sigma methodology to problems that they encounter in their regular work. In some cases Green Belts may move on to become Black Belts but only in the case where there is an organizational need and the candidate meets the organizations requirements for the position. Occasionally a Green Belt may be recruited to be a member of a Black Belts larger scale cross-functional project team. Typical Green Belt projects yield cost savings/avoidance in the range of US$25,000 to US$100,000 and are completed in 2 to 6 Months.

Yellow Belts are an important part of the Six Sigma infrastructure. In the early days of Six Sigma implementation, Black Belts found themselves spending a lot of their time teaching needed project team members the basic language and roles of Six Sigma. Items taught included the roadmaps (DMAIC/DMADV), basic data collection and display, Pareto Principles, Mapping, etc. When these early Black Belts were debriefed to learn what would make the process more efficient, most said that some basic level of training for team members prior to working on a Six Sigma project, would enable them to focus more fully on the project as opposed to training. This speeds up project cycle time and allows for a more efficient use of resources. Yellow Belts usually are from a wide range of the organization that either need a basic understanding of Six Sigma, (including members of Management and Champions) or are slated to be a team member on a Six Sigma project. One ancillary benefit of this type of training is that the Six Sigma methodology, metrics, language and approach are driven deeply into the organization and culture. In organizations where this has occurred Six Sigma yields the highest results.

Deployment Roadmap

The following illustration depicts a typical Six Sigma deployment roadmap. It is important to note the order in which training occurs so that each layer of the organization is trained, prepared and is practicing their specific roles in a timed sequence designed to maximize results and eliminate risk of failure.

Figure 2: Typical Deployment Wave

This deployment example will yield 10-20 Black Belts and 10-20 Green Belts all of whom will complete a project during the training sequence. In addition, 20-30 Yellow Belts will be trained in the foundations materials to support the projects as needed. This deployment would be typical for a small organization with 1000 employees or would be a good Phase 1 model for a larger deployment where multiple training waves would be required. During the training phase a typical cost savings / avoidance of between $1.25 Million and $6 Million (depending on the number of students, usually 10-20, and the ranges of savings mentioned earlier in the article) would be realized. Once all candidates are certified and the complete system is institutionalized, annualized cost savings/avoidance from this size of deployment would be between $4.5 Million and $20 Million per year, again based on the number of Black Belt and Green Belt candidates and the per project savings thresholds achieved.

Reaction to these types of numbers is often skeptical but the savings ranges are based on a long history of thousands of projects and hundreds of companies. The early adopters in software have already validated that these ranges are in fact available and achievable. Six Sigma for Software and IT is now starting to gain momentum and with the current expansion of companies moving towards Six Sigma in this environment the number of success stories and the availability of case studies will further substantiate this model and it’s associated savings figures. Through this forum we will continue to bring that information forward and invite others to share their successes, idea’s questions and builds.