Most of the comments in this column are based on information from [2], [5], [6], [7], and [8].  These references are not meant to be an all-inclusive list but rather a representation of some of the material currently available for reference.

In an earlier column,^[1] the basic concepts of modeling in Six Sigma projects using a technique called SIPOC were discussed.  The conclusion was that, as a singular modeling method, Six Sigma's SIPOC is not sufficient to understand all the aspects of a process in order to define, accurately analyze, and suggest improvements to processes.  However, it can be somewhat useful as a "high level" view of a process.  This, of course, assumes that the process itself has been correctly defined.

The purpose of this column is to explore the more detailed concepts surrounding Six Sigma and to determine whether it's a sufficient approach in the understanding and improvement of business processes in service organizations.  Figure 1, the SIPOC Diagram, represents the basic concepts of the SIPOC notation.

Figure 1.  Basic Concepts of the SIPOC Notation

In order to evaluate the usefulness and appropriateness of applying Six Sigma to the improvement of service processes, it's essential to understand some of the basic concepts of that discipline.  In its purest form, Six Sigma is all about eliminating defects from a manufactured product.  The approach taken to do this is comprised of five phases (or steps or activities).  These are summed up in the acronym DMAIC.  DMAIC stands for Define, Measure, Analyze, Improve, and Control.^[2]

• Define the problem and what the customers require.
• Measure the defects and process operation.
• Analyze the data and discover causes of the problem.
• Improve the process to remove causes of defects.
• Control the process to make sure defects don't recur.

In organizations where any kind of structure for approaching process improvement is missing, these activities probably make a lot of sense.  In fact, if combined with other concepts and approaches, there is nothing wrong with these activities.  The question to answer is, "Are they sufficient and appropriate for service organizations?"

There is nothing wrong with the concept of near-perfect products and services for customers.  The purpose of this column is to explore the method/framework/discipline/approach called "Six Sigma" applied to Service Sector organizations and organizations primarily providing services, in contrast to organizations primarily providing products.  As I pointed out in "The Invisible Process"^[3] and "Is It Cake or Is It Clean?"^[4] manufacturing and service processes are different and should be treated differently.  Concepts that make perfect sense in a manufacturing environment may not even exist in a service environment.  For example, many of the metrics and calculations that are so much a part of the Six Sigma method are not even captured or tracked in a somewhat similar service process.

In "SIPOC for Service — Is it Enough?"^[1] the discussion focused only on the SIPOC notation and not on any of the other aspects of Six Sigma.  That discussion pointed out that the lack of attention to the concepts of 'guides' and 'enablers' causes concern since without this information an organization is missing about 80% of the data needed to make the necessary decisions related to project definition and problem identification.

I've met and spoken with several people who either are Black Belts or worked in organizations that had several Black Belts.  Those that worked for service organizations, without exception, said that Six Sigma, as it is structured for success in manufacturing, didn't work for them.  They were able to take some concepts from Six Sigma and mix them together with concepts from other business process management frameworks to create success stories.

Six Sigma:  The Breakthrough Management Strategy^[5] argues that it isn't Six Sigma that creates the successes; it's the combination of some of the elements of Six Sigma with some really good process management strategies like executive buy-in, dedicating a significant budget to training, and tying the results from projects to executive compensation.  Any really good framework or methodology will (should) incorporate these concepts.

LEAN Six Sigma for Service^[6] discusses the incorporation of the valuable LEAN concepts into Six Sigma, again taking it further and incorporating some good process management concepts that have created the success stories told in this author's writings.  In addition, in "Six Sigma for Financial Services"^[7] companies like American Express, Bank of America, Wachovia, and others were not implementing just Six Sigma but also utilizing some concepts in conjunction with LEAN along with some good process management methods.

The purpose of this column is not to question the usefulness of Six Sigma but to create a discussion around whether or not it alone is a sufficient choice for organizations in the service sector.  Taking a look at each of the concerns below, we can deconstruct each of the success stories for Six Sigma and see how they became successes.

There are a multitude of concepts surrounding Six Sigma that make it a questionable approach for service processes.  They include:

• Its focus is on 'defects' to the exclusion of other aspects of a service process that are critical to success.
• It assumes that control points are in place and that defects only occur at the beginning or end of the process (based on types of metrics and when they are collected).
• It was designed specifically for a manufacturing environment.
• It focuses on activities or pieces of processes instead of the whole process or value chain.
• It ignores the concepts around adaptability/flexibility.
• It doesn't focus on any stakeholder other than 'Customer'.
• It ignores the people aspect of process.

It ignores the 'guide' and 'enabler' concepts, e.g., policy, procedures, regulations, knowledge, experience, and events through the use of the SIPOC modeling method.

Focus on Defects

Six Sigma, what is it?  Sigma in Greek stands for "standard deviation."  Six Sigma is 99.9997% perfection.  The main focus of a Six Sigma discipline is on statistics, not "gut-feel."  It's really all about how to collect and analyze statistics to determine the defect rate of whatever is being measured.^[8]  In a manufacturing environment, it's much easier to identify and measure defects.  When following Six Sigma, most of this information would be gathered from efforts called "voice of the customer."

However, in a service environment it's much more difficult to identify what a 'defect' is and to measure how much it matters to the customer (or that we "feel" it matters to the customer).  In "Is it Cake or is it Clean?" I discussed this dilemma.  In short, when you are making a cake, there are very well-defined criteria for whether it's a quality cake or, in fact, if it contains defects.  But when you're cleaning something (like your house) how do you define 'clean'?  How do you quantify clean for the cleaner?  How do you define the defects of clean? 

This has become much more difficult because what we're trying to define is an intangible concept.  My definition of clean and my house cleaner's definition of clean may not be the same ... and frequently are not.  Lately, I've been giving the job of cleaning the house to my thirteen year-old daughter and have realized that her definition of clean is also different from mine and that of the cleaning service.  Some of this is simply due to a lack of experience.  But a lot of it is based on perception and acceptance criteria and effort involved. 

What I have come to realize is that, when trying to define defects, it depends on who the customer is.  Most service organizations put a lot of effort into gathering statistics for customer satisfaction, but 'customer satisfaction' can vary depending on the customer's last experience.  The company could be providing the same level of service and some other variable has created a negative experience for the customer.  So, if the company were trying to collect defect data, would there be a defect in the service or the circumstance?

The dilemma of the intangible service makes it frustrating, if not impossible, to define and collect data for defects.  That is not to suggest that service organizations don't need to understand this concept, but the reality of implementing a true Six Sigma calculation may cost more than the return on the investment.  That takes us back to the Quality Dilemma of the '80s and '90s where quality programs were found to not be cost effective.

Organizations that attempt the Six Sigma approach for their services find themselves forced to focus only on the tangible aspects of the process such as transaction volumes, error rates, re-work, and time consumed.  The danger is in the intangibles, such as ability to adapt and change, to respond to changing customer expectations, etc.  It's the inadaptability of service processes that is hurting many service organizations, and yet this is one area that Six Sigma, in its truest form, ignores completely.

For those organizations that provide services, what makes them successful is not how few defects they produce but rather (on a more positive note) the number of things they do right.  Staying focused on doing things right and doing them better will eventually eliminate the defects.

Specifically Designed for Manufacturing

Manufacturing-based organizations desperately needed something in the late '80s and '90s to fix the mistakes of the '70s that had almost completely devastated the manufacturing industry in the U.S.  Motorola was one of the first to move away from the traditional approach to solving manufacturing problems that invested large amounts of money into yet another quality program that might or might not work.  Instead, they started looking outward to the customer and inward to their own processes to resolve their quality issues.  What Six Sigma provided Motorola was an attainable, very aggressive quality goal of almost defect-free products.  The metrics for this were easily attainable and fairly easy to track and collect because of the manufacturing environment.

So, other companies — recognizing that the success at Motorola was accomplished without an overinvestment in the "Japanese" type quality programs — jumped on the bandwagon, eager to replicate Motorola's success.  GE (a mixed organization that contained both the manufacturing environment and the service environment) jumped in with both feet and, as a result, was also able to attain some major successes.  However, at GE they were already finding that they needed to incorporate some significant changes to the Motorola approach.  The most significant change was tying the results to executive compensation.  Other aspects of the GE approach included:^[5]

• In addition to creating a monetary incentive, CEO Jack Welch also told management that if they wanted to be promoted to a senior management position within GE, they must start Black-Belt (and even Green Belt) training by January 1, 1998.
  
  
• GE already had a program in place called "work-out" that created an environment within the company for employees to question "the way things work."
  
  
• GE invested $250 million in training and trained over 64,000 out of a total workforce of 222,000.  This created an environment where a large portion of the workforce had a very clear understanding of the Six Sigma concepts.  Such an understanding is critical to the success of any process improvement project, whatever method is used.

In summary, some would say that GE did everything right.  But it wasn't Six Sigma that created the success; it was the management support and buy-in, the training, and the push from the very top of the organization that made this successful.   I would counter that there are a number of other, less well-known frameworks that could be just as successful from an improvement perspective and, in fact, cost much less, resulting in a much higher ROI.  The Six Sigma training is some of the most expensive of its kind. 

What made these success stories was not Six Sigma as it's strictly applied in manufacturing but rather a collection of other common sense factors.

Measures for Six Sigma

In discussions with Black Belts and with further research, I found it very interesting that the entire focus of a Six Sigma project was on effectiveness and efficiency.  I could not find anything that suggested organizations should try to collect data on and measure adaptability.  Almost all of my consulting work is with service organizations, and in the last five years I haven't had one organization that didn't list adaptability or flexibility as one of their major concerns regarding the services they provided.  They readily recognize that when they are beaten by their competition, it's because they were too slow to respond to the market.

In addition, the granularity of the measures needed to accurately calculate Sigma are not readily available in service type organizations, and it may not be feasible to collect when the cost of collecting them is factored into the project.

Six Sigma as an Activity vs. Process Focus

Most organizations approach Six Sigma as a problem-solving approach, rather than as strategic improvement or business transformation.^[8]  This can prove to be a very disruptive approach, particularly in a service organization.  In providing services it's more likely for processes to cross several organizational boundaries.  Therefore, an approach that focuses on solving problems in just a single area/function in an organization can cause many more problems than it solves because it is simply moving the problem from one part of the organization to another without really eliminating it or discovering its cause.

This also makes change management much more difficult because it addresses change in small increments, in isolated parts of the organization.  Change management experts will quickly point out that "real" change management is only effective when it encompasses as much of the culture of the organization as is feasible.  For example, GE was successful in its Six Sigma projects because it was a cultural change that crossed all boundaries of the organization.  GE already had a program called "work-out" in place that crossed many of the normal barriers that cause problems in other companies.^[5]

In order to be truly successful, a process improvement project must follow this example.  It needs to look at the whole process, not just small portions of it, applying common "fire fighting" techniques.  The fire might be extinguished but nothing has addressed how to prevent it from occurring again in the future or from breaking out in a different part of the organization.

It's only when LEAN concepts are incorporated with Six Sigma that the idea of the "value chain" is considered.  Many, if not all, of the success stories in LEAN Six Sigma for Service can be attributed to the fact that the focus of the improvement project is on a much broader view of the process.

The "Voice of the Customer" (VOC) is the Focus of Everything

In a Six Sigma project, the "voice of the customer" is everything.  It's used to define defects and to determine project definitions.  Most organizations probably don't give enough attention to the "voice of the customer."  However, there are numerous other stakeholders in an organization that also impact success.  For example, other external stakeholders include suppliers, financial institutions, government regulators, community, and competitors.  There are also internal stakeholders that should be considered, such as other processes, business units, and staff.

The idea that projects are chosen based almost solely on the "voice of the customer" is concerning.  What happens to the expectations of all the other stakeholder groups?  What opportunities have been missed due to this focus?  All stakeholder expectations should be understood and factored into the decision.  Not all stakeholders carry the same weight when these evaluations are made, but it is important that an organization understands the impact to those stakeholders.

In addition, the idea of how the work of a project "aligns" to all the areas of an organization is also important.  Consideration should be given to whether the perceived problem to be solved is the result of a poor alignment of technology to organization structure, or organization structure to strategy or knowledge or perhaps even capabilities.  Sometimes process problems are all about capabilities.  Given a different set of capabilities in the workforce, the process problem might be 90% solved.  In a Six Sigma project it is highly likely this issue might not ever be identified, resulting in continuous fire fighting.

What Makes Six Sigma Seem Successful?

The author of Six Sigma:  The Breakthrough Management Strategy suggests that the following elements create a successful implementation of Six Sigma:^[5]

1. Highly-visible, top-down management commitment to the initiative;
2. A measurement system (metrics) to track progress;
3. Internal and External Benchmarking of the organization's products, services, and processes;
4. Stretch goals to focus people on changing the process by which the work gets done, rather than "tweaking" the existing process;
5. Educating all levels of the organization;
6. Success stories to demonstrate how the approach is applied and the results;
7. Champions and Black Belts to promote the initiatives and to provide the necessary planning, teaching, coaching, and consulting at all levels of the organization.

There's actually nothing wrong with anything this author states.  The problem is that none of it is part of the Six Sigma approach.  These are all the things that need to happen outside of the concepts of Six Sigma, and it's inappropriate to attribute an organization's success with process improvement to Six Sigma under these circumstances.  That success is due to the implementation of process change supported by all the elements mentioned above, not due to anything that Six Sigma specifically brings to the project.

Conclusion

In conclusion, after discussions with Black Belts and further research into the topic of Six Sigma, it seems apparent that it is not the discipline of using the Six Sigma defects calculation or the DMAIC framework that helps organizations, especially service organizations, make those really incredible improvements to process.  It calls for a much broader approach than Six Sigma, such as one that incorporates LEAN, which considers the whole value chain and not just pieces and parts of it.

There are a multitude of process management concepts that suggest that it is important for an approach to include things like process architectures and alignment of the organization and the expectations of other stakeholders in addition to the customer.  Incorporating executive compensation as part of the motivation is a tremendous critical success factor.  There is a saying that "we get what we pay for."  It is apparent that, if executives are motivated and the organization invests literally millions of dollars into training its people, the likelihood that something in an organization will change is dramatically increased.  However, none of that happens if an organization is implementing just a Six Sigma framework.

This column has examined concepts of Six Sigma that make it a questionable approach for service processes.  There seems to be a legitimate concern for trying to apply Six Sigma concepts in service organizations.  Morphing something that was developed and intended for use in a manufacturing environment — where there are tangible products to monitor for defects and where measures are collected constantly — into something to use in service organizations — where almost everything that happens is intangible and few (if any) measures exist — seems like a lot of hard work for very little, or no, gain. 

A better approach for a service organization is to adopt a good process management framework that starts with the strategy of the organization, defines all the stakeholders and their expectations, and then aligns all of that with all aspects of the organization (such as organization structure, technology, knowledge, capabilities, etc.).  Then, the organization can develop a definition of its processes and conduct an assessment of the health of those processes.  Next comes priority setting on which processes will provide the greatest return (ROI) when improved and a program management strategy set up to manage those projects.  Apply the common sense approach of defining the project, understanding the current situation, capturing measures to support those conclusions, and analyzing the process using root cause and value-add techniques.  Then, based on the results of all of this work, improvements to the processes can be proposed and implemented, the progress of the implementation monitored, and any corrections made where necessary.

In the final analysis, Six Sigma does seem to bring a discipline to those that have none.  The basic approach of Six Sigma — define, measure, analyze, improve, and control — should be applied to every process improvement project.  The larger question is, "Is it sufficient?"  The conclusion is "no" — not in service organizations.

References

[1]  Kathy A. Long, "SIPOC for Service — Is It Enough?" Business Rules Journal, Vol. 11, No. 9 (Sep. 2010), URL:  http://www.BRCommunity.com/a2010/b553.html

[2]  Peter S. Pande, Robert P. Neuman, and Roland R. Cavanagh, "The Six Sigma Way," Team Fieldbook, New York:  McGraw-Hill (2002).

[3]  Kathy A. Long, "The 'Invisible' Process  (Where are Your Processes?)," Business Rules Journal, Vol. 10, No. 9 (Sep. 2009), URL:  http://www.BRCommunity.com/a2009/b498.html

[4]  Kathy A. Long, "Is it a Cake or is it Clean?  (A Litmus Test for Process)," Business Rules Journal, Vol. 10, No. 11 (Nov. 2009), URL:  http://www.BRCommunity.com/a2009/b508.html

[5]  Mikel Harry and Richard Schroeder,  Six Sigma:  The Breakthrough Management Strategy Revolutionizing the World’s Top Corporations, New York:  Doubleday (2000).

[6]  Michael L. George, LEAN Six Sigma for Service, New York:  McGraw-Hill (2003).

[7]  Rowland Hayler and Michaek D. Nichols, Six Sigma for Financial Services, New York:  McGraw-Hill (2007).  

[8]  Peter S. Pande and Larry Holpp, What is Six Sigma? New York:  McGraw-Hill (2002).  

# # #