Where was I? Oh, now I remember! All the talk about Six Sigma… it’s the latest thing, it’s just a bunch of old techniques, it can perform miracles, it can do a lot of damage if you’re not careful. There’s a lot of meaningless chatter on this subject, but fear not… I’m here to help you! When you sit back and put all the information together, the answer is pretty clear. Of course to fully understand the issue you need to also read yesterday’s Blog, but when you but the two together I know that you’ll agree with the conclusion I’ve reached. And what is that conclusion? Simply put, you NEED to question if Six Sigma is right for your operations! Let’s take a closer look, and you’ll see why!
Every system or philosophy in existence has problems. Your current operation has problems. The real question is, “Will a new system work in my environment?” In almost every case, Six Sigma only woks in PART of your environment. Why? Well, the problem is built into the name… Six Sigma. That means a goal of 99.999 perfect. While improvement is a good goal, this goal is neither realistic nor desirable in a service organization. When Motorola created Six Sigma, the organization was already producing at 3 Sigmas (99.7% perfect). In corporate service groups, some management reports may show performance levels of 95%, but most of these reports are… unfortunately… wrong. Real service levels are somewhere in the 60%-80% level. Scattered around you may find a few services that truly perform at 97% or 98%, but they are very rare and they may not really be true “service” groups”.
The difference between an industrial function and a service function (at least for this discussion) is the role of the client is in the production process. To understand this, let’s look at one of the success stories of Six Sigma… microchip manufacturing. The microchips in our computers (CPU, RAM, etc.) are marvels of modern industry. In a device that’s smaller than your fingernail, there could be more than a Billion transistors, and every single transistor needs to work reliably for the life of the chip. This industry couldn’t exist at Three Sigma; every chip would have so many flaws that they would not function. When Motorola began to look into Six Sigma, typical microchips had a few thousand transistors. By the next decade, chips had moved to a million transistors. In order to increase the complexity of a product by 1000 times and still have if function reliably, an equally massive improvement in quality (reduction of errors) also needs to take place. To do this, absolute control needed to be imposed on the production environment. These special places are called “clean rooms”, with the air specially filtered to remove any particles, a system of airlocks to get into and out of these rooms, and the personnel in “bunny suits” (full body gowns, masks and hairnets). Pencils are forbidden (a particle of graphite could disrupt the day’s production) and everyone must speak softly or the most delicate equipment cannot be fully calibrated. Does this sound like your production environment? Are you trying to block the client from access to the service process? For most client services the client is clearly “inside” the production facility. Bankers and Lawyers modify documents even while they are being turned through the document center. Computer support groups go to user desks and wait (hope?) for the user to return so they can access their laptop. A corporate library often has to negotiate with end users to guess the identity of the firm they are researching (ex.: a senior executive hands off a request to a junior executive who passes it to a secretary who places the request… losing data at every step). In a service the client MUST be part of the process. And neither actually nor metaphorically are any of us going to get our client into a bunny suit.
Surely the literature on Six Sigma must have something to say about this! And it does. It says, “No system can be Six Sigma unless all parts are Six Sigma.” Let’s go back to our chip plant. Having a clean room is just one part of the process. When they create silicon chips, they need absolutely pure silicon as a base ingredient. Suppliers who couldn’t meet the purity requirements were dropped. Unexpected impurities in the silicon would generate faulty chips, the laser etching devices must make incredibly precise cuts for the chips to work correctly, adhesives holding chips onto circuit boards must resist daunting heat conditions, and on and on for every element in the production process. Let just one element slip and quality may drop by orders of magnitude. In the industrial world, the Procurement department plays a pivotal role in converting manufacturing needs into vendor relationships and contractual requirements. In the service world, this is impossible. In E-Discovery, the process of reviewing documents related to a lawsuit requires documents to be provided by the client, adherence to court schedules (which a judge can, and will, change at any time), re-scoping of the number of documents under review (by the lawyers, by the courts, because of changes in legal jurisdiction, etc.). None of these agents are under the control of Procurement, nor are they paid by the “manufacturer” and subject to replacement, leaving services with few levers over the materials used in production.
One last, but vital, point. An industrial firm is highly dependent on the use of very specialized equipment. This equipment becomes more sophisticated with every generation, usually incorporating all the features of the previous generation plus new features, higher speed and a lower price. Because the equipment is owned the manufacturer, they will continue to receive this generation of equipment until it makes sense to replace it with a newer generation. Computers, for example, double their speed about every two years. Services, while they may use computers or phones or other relatively generic technologies has many more elements of their processes carried out by people. With a large number of the processes human dependent, rather than machine dependent, the rate of improvement is unlikely to meet the “10 time reduction in errors every 2 years”… the standard for Six Sigma. Also, machines do not walk out the door and take their embedded business processes with them. We can capture best practices and train new staff (or at least some of them) up to the level of the best staff that left, but the training treadmill counts as “waste”, whereas replacing a five year old machine with a new machine that is half the price and can produce 10 times the product, is “value”.
It’s not really a single system. It is a collection of processes that has been refined and expanded over time, bases on what worked and what doesn’t work. In fact, then next Six Sigma specialist you work with was probably trained in Lean Six Sigma. By incorporating Lean methodology, it addresses more projects that are more focused on speeding up turnaround than reducing costs. As time goes by, and Six Sigma moves deeper into service delivery (instead of just industrial processes) it will probably incorporate additional tools and might even break into “Six Sigma for Services” or “Six Sigma for IT” or other more specialized versions. What have we learned?
- Six Sigma provide tools that can benefit a service environment, but it’s industrial origins set goals and objectives that are unrealistic, even inappropriate, for corporate service.
- Best in class industrial production requires an increasingly isolated, inflexible and tightly controlled environment; best in class services incorporate the client’s open environment, require flexible processes and changes to the production process to accommodate the changing needs of clients.
- Six Sigma uses Procurement to eliminate vendors which cannot meet Six Sigma standards, thus avoiding new errors and maintaining quality. Inputs to service processes come from clients, customers, regulators, and other sources that are not paid and therefore, not subject to the Procurement process.
- Industrial processes are primarily performed by equipment, which usually offers orders of magnitude of additional value when replaced. Service processes are primarily performed by people, which not only offer relatively limited room for improvement when they are replaced, but replacements may initially offer less value than their predecessors (until they are trained in the new environment). Equipment replacement offers an immediate increase in value, people replacement entails hiring and re-training costs, generating waste.
In the end, is there a value to Six Sigma in service organizations? Yes, there is. Can anyone point out a Fortune 500 service with Six Sigma quality levels (OK, 5 Sigmas? Anyone have 4 Sigmas?)? Most services are battling to get to 3 Sigmas, and winning that battle may produce a service with high technical performance, that mysteriously fails to improve client satisfaction. We shouldn’t throw out what’s good about Six Sigma just because it has flaws, but we need to be ready to adjust… perhaps rethink… the targets and goals of Six Sigma. It’s like dividing the circumference of a circle by its diameter. Technically it’s 3.141592… followed by an infinite number of digits. But for most real world purposes it makes more sense to just use 3.142. We need a Services version of Six Sigma that is not as complicated as Pi. And that’s my Niccolls worth for today!
I’ve caught up with all your posts since Feb ’11. Hope everything keeps going well with your blog, I like the name and sense of fun within the posts. Freakonomics was a truly excellent book but Superfreakonomics not so exciting but worth the read.
You last 2 blogs have reached into my area (@Lean_Consultant) of focus, although we specialise in industrial use as many within the Lean Six Sigma communities do. I have been lucky to apply a specific part of Lean within services for a FTSE 100 company to critical acclaim. Back in 2002, a group provided engineering services within the European Division of this FTSE 100 company was also responsible for deploying Total Productive Maintenance (TPM – not Tea Party Movement) across 50 or so factories.
We were led to understand by our Japanese Institute for Plant Maintenance (JIPM) assessor, Kunio Shirose, that this department was the first outside Japan to receive the JIPM Excellence Award for application within the Office Environment in 2002. Even after our successful application, our service level was not Six Sigma (far from it!) but we had started on the journey which could eventually lead us to success. Within TPM the concept of ZERO is very important, and as a subset within Lean this adoption of ZERO continues. The ZERO represents a tenacity to believe and push all efforts towards the elimination of losses (deviation from Standard) and wastes (incorrect design/content of Standard).
In my opinion, it’s just that, the “bunny suits” for your clients are not the wishes they have to make immediate and inconvenient changes, but the system that controls the change. As long as there is a “standard” way of working with the client, the problems (both deviation from standard and inherent problems of the standard) can be found and eliminated.
Good luck on your search for Six Sigma in IT and Services, I’ll keep reading to be kept informed. My last opinion is “remember that the simple Lean Leadership tools provide enough problem solving capability before using any of the sexy tools of Six Sigma ones”.
Great to hear from you: few clients are trying to disrupt production, they are just reacting to the system they have to work with. I also agree that simple is better. The basic tools of Six Sigma will take you pretty far… probably farther in corporations because it’s such a “green field” for any quality initiative.
Over the next few months I want to pull together opinions from around the web (and the world) on the critical differences between Industrial and Corporate practice in applying Six Sigma and Lean. I suspect that this specialization could provide similar benefits to what we are seeing in the medical field.
Look at Pediatric medicine. In the past, doctors treated children like small adults. Generally this works well enough, but it leaves much to be desired for pediatric cancer treatment. Less than a generation ago survival rates were around 35%, but now it is over 80%. According to one study, the big didn’t come from the expected sources (new drugs, improved radiation therapy, or other new “cures”). Instead the big change came from a social initiative. A group of pediatric oncologists agreed to regularly meet and share best practices. And success rates skyrocketed! Will the same apply to Corporate Six Sigma… we’ll see!