Six Sigma : Business Metrics

[sk_kireeti]

OVERVIEW

“When you can measure what you are speaking about, and express it in numbers, you know something about it; but when you cannot measure it, when you cannot express it in numbers, your knowledge is of a meagre and unsatisfactory kind. It might be the beginning of knowledge, but you have scarcely, in your thoughts, advanced to the stage of science.”
—WILLIAM THOMSON, LORD KELVIN (1824–1907)

If you don’t have measurements, you can’t progress intelligently and efficiently because you don’t know where you are. Quite simply, we need ways to keep score. In business, the most important scorecard is profit. However, money is not a sufficient metric in itself because it leads businesses to be penny-wise and pound-foolish.
A business metric is any unit of measurement that provides a way to objectively quantify a process in ways that make sense and dollars for your organization. Any measurement that helps managers understand their operations might be a business metric.
Most businesses have some version of measurement. After all, managers and executives need to calculate profits and losses, the cost of goods or services sold, and return on investment. But beyond the basics, how exactly do managers go about making decisions and changes that reduce costs, improve profitability, and foster growth?
Many managers simply accept and communicate and operate according to certain beliefs that they hold as truths. However, when managers are pressed to objectively justify their beliefs and explain how they provide appropriate guidance, they’re often at a loss. And, as quality guru Edwards Deming said, “Without data, all anyone has is an opinion.”
Metrics should be a fundamental part of running any business. Measurement is a fundamental value and belief of Six Sigma.
Every Six Sigma project depends on metrics, whether you call the set of metrics your scorecard or your dashboard or whatever. Improving processes— whether for products or services or operations—requires improving how we do things. This means looking at how we track what we’re doing.
Business metrics provide data that Six Sigma managers can use to better understand their processes and identify target areas for improvement. If you can measure your processes, you can understand them. If you can understand them, you can correct, control, and improve them and thus reduce costs while improving the quality of your outputs.
Six Sigma metrics might include customer satisfaction, percent of defects from a process, number of products completed per hour, hours required to deliver a certain number of outputs or provide a service, the cost of poor quality, the cost of goods or services sold, and so on. These metrics quantify the effects of those factors that matter most.
How do we measure our progress? Are those metrics appropriate? Are they effective? Before you undertake any Six Sigma project, you must ask and answer the big question: what should we measure?
IT ALL STARTS WITH THE CUSTOMERS

“The best Six Sigma projects begin not inside the business but outside it, focused on answering the question—how can we make the customer more competitive? What is critical to the customer’s success? . . . One thing we have discovered with certainty is that anything we do that makes the customer more successful inevitably results in a financial return for us.”
—JACK WELCH

What do your customers want and expect from you? How do your customers understand quality?

The answers to those questions provide the basis of any Six Sigma project. The first step toward improving processes, products, or services is determining what your customers consider critical to quality. The critical-to-quality (CTQ) concept in Six Sigma allows you to focus on improving quality from the perspective of the customers. This is often called “voice of the customer” (VOC). VOC confirms and validates the CTQ, or in some cases, the VOC might change the CTQ.


Managers and employees all have some ideas about what constitutes quality for their products and services. That’s good—but it’s not what matters most, because managers and employees don’t put cash in the coffers. Find out which aspects of your products and services are vital to the customer and in what ways. Then managers and employees can devote themselves to finding ways to improve those products and services by improving the processes behind them.
Here’s a quick and simple example. Most of us have used the drive-through window at McDonald’s. So, as a customer, what would you consider the CTQs for McDonald’s drive-through process? How long it takes to deliver your order? Whether the order is correct? Whether the Happy Meal contains a toy? We know that taste is not a CTQ for most normal adults going through the McDonald’s drive-through. We know that it’s not the hot spot to take a date. So, if you have to wait long for your order, that’s a defect. If the employees get your order wrong, that’s a defect. If the Happy Meal is missing the toy, that’s a defect.
Now, the manager and the employees might have ideas about how they would like to improve the drive-through process, but what matters most is what you and the other customers consider important. Those CTQs should be the basis of any improvement efforts. Those CTQs are the terms by which customers understand defects. The defects defined by those CTQs should be what managers focus on analyzing, to identify and reduce or eliminate the factors that cause those defects. A general guideline to set CTQs are by customer expectations that can be measured both financially and at internal/external defect levels.
Here’s another example. At a conference in a hotel, I asked participants what they expected in their coffee breaks. The answer: “lots of good, hot coffee!” When I asked the hotel banquet staff what they needed to provide, they agreed on good, hot coffee. But the two groups differed in their CTQ expectations.
Beyond coffee, the staff was concerned with providing linens, china, attractive displays, and extra snacks. However, the customers wanted a fast line for refills, high-capacity restrooms nearby, and access to telephones. Of course, customers don’t want dirty cups or grubby linens, but they don’t care much about ice sculptures.
So, here’s the bottom line. The hotel is putting time and money into things that matter less to the customers and missing out on things that customers expect. Although the English author and diplomat James Bryce was not referring to Six Sigma, his comment seems very apt here: “Three-fourths of the mistakes a man makes are made because he does not really know what he thinks he knows.”

CRITERIA FOR BUSINESS METRICS

So what are the criteria for establishing business metrics? The answer is surprisingly simple. The criteria relate to why you’re in business and why you want to implement Six Sigma—to improve customer satisfaction and reduce costs. With that in mind, you need to establish metrics to help you achieve these goals.
Measurement is crucial to the success of your Six Sigma initiative. Your business metrics show you the ways to achieve dramatic improvements in your processes. They apply statistical tools to any process to evaluate and quantify its performance. They continually ask what outcome, or dependent variable, is a function of another, independent variable, to dig out the information that improves your performance.
It’s a fundamental principle of the Six Sigma philosophy that you cannot improve quality unless you can measure it. This applies to any aspect of Six Sigma. If you’re going to invest in measuring customer satisfaction, then you need to have a quality measurement system to track performance. In other words, your business metrics constitute your scorecard, the way you figure out where you are. To use another term, they form your dashboard.
The Dashboard Concept

The business dashboard as a metaphor for critical metrics to measure business performance originated years ago at General Electric. Just as you use the speedometer, oil gauge, battery indicator, fuel gauge, and other instruments to monitor the status of your vehicle as you drive, so you want to keep track of key indicators of the performance of your company. Like the dashboard gauges, your metrics allow you to continually assess your progress and detect any potential problems.


Your choice of business metrics and the importance you give them show what you value. If you value customer service, for example, you make it central to your business metrics. You put it on your scorecard. You include it in your dashboard.


Conversely, if you don’t measure quality and don’t follow up on any measurements, you give the impression that you don’t care about customers or profitability. That impression affects the behavior and productivity of everyone in the department, division, or organization.
By instituting key business metrics across functions and groups and at every level, you directly link individual performance to measurable outcomes. This sends a clear message that not only do you care about customers and revenue, but so should all team members because they are accountable for the results measured by their particular metrics. In the end, what we are talking about is establishing the proper “dashboard” of measures to help us understand our quality situation and to create urgency to take the actions needed to address the root issue: defects.
QUESTIONS ARE KEY

When you create metrics, you need to ask questions—new questions—and search for new results. If you keep asking the same questions, you’ll keep generating the same measurements. That brings to mind that adage: “The height of insanity is doing things the same way and hoping for a different result.” How many companies exemplify that insanity in managing their processes?
How do new measures arise? Fundamentally, new measures arise when new questions are asked.
If we ask questions about outputs, not inputs, we get a lot of focus and measures on outputs. If we get questions on budgets, we measure budgets. How many of our questions are focused on process and product quality? How many of our questions are focused on causes and inputs versus results and outputs?
Questions are key. To get new metrics—and new behaviors—we must ask new questions.
There’s another quote to keep in mind:

“Genius, in truth, means little more than the faculty of perceiving in an unhabitual way.”
–WILLIAM JAMES (1842–1910)

Six Sigma allows and even requires you to be a genius. The key to “perceiving in an unhabitual way” is asking questions––and then questioning the answers.
When you begin setting up new business metrics and asking questions, you should begin with the fundamentals:

• Why do we measure this?
• Why do we measure it in this way?
• What does this measurement mean?
• Why is this measurement important?

Ask questions. Challenge answers. Put assumptions to the test. Confront conventions. At this stage you should tap people who are known for their critical thinking skills, whether they’re familiar with the processes or not. Encourage them to question and to challenge. That approach might be unusual in your company, but to do otherwise would be the height of insanity.
Another behavior that’s insane is to make decisions without using metrics. Sometimes managers trust their instincts or play their hunches or just guess at what might be wrong and how to solve the problems suspected. But why would you simply guess at ways to improve products, satisfy customers, and increase revenue? Sure, you might find a way to reduce expenses in a given process without using metrics. But how will the change affect customer satisfaction? Are there ways to improve the process as well as cut costs? Is the process even necessary?
Measure first and measure often. It’s the best way to improve processes, meet customer expectations, and reduce costs. And, to offer one more quote, artist Edward Simmons said, “The difference between failure and success is doing a thing nearly right and doing it exactly right.”
LEARN FROM YOUR CUSTOMERS

Here’s how you can best proceed to get the most from your Six Sigma initiative. From metrics to project selection, as long as you follow the basic principles expressed in the following list, you can ensure that your efforts are working at maximum capacity for lasting results.
Start to ask questions—different questions—and really desire to seek out data that counts. Use the simple customer-based management survey tool to promote awareness. Everything yo do or produce affects more than one customer. Who are your customers and what are their top three requirements?

Figure: Customer-based management survey


If your company is like most others, this process will clearly show that you are stuck in four sigma or less, along with most of your competitors. You are just average—and that is nothing to celebrate. Especially when you realize that one of your new competitors is in the Six Sigma group of world-class companies. You might suddenly understand why this competitor is giving your sales manager such difficulties in the marketplace. They aren’t dumping their product at an artificially low price. Instead, they are able to underprice you because the highest-quality provider of goods and services is the lowest-cost provider.

[sk_kireeti]

MOVING FROM CRITERIA TO METRICS

How do you actually select appropriate metrics? You need to measure what’s important, what’s critical to your business. You know that key criteria are customer satisfaction and revenue. So, let’s start with customer satisfaction. For every product or service, you need to determine the expectations of your customers, particularly the CTQ factors. What aspects of the product or service are key to your customers? For each aspect, what are your customers’ expectations?
If you’re producing widgets, for example, you might consider the following aspects: size, weight, durability, price, ease of use, versatility, colors, styles, availability, maintenance, service, warranty, and so on. For each of these aspects, you would determine expectations. How many sizes of widgets do they expect? Which sizes do they expect? How long do they expect a widget to last? What price do they expect to pay? How easy to use do they expect a widget to be? How long do they expect it to take to repair a widget? Do they expect a loaner widget?
For every product or service, there could be dozens or even hundreds of aspects. Focus on the aspects that are most important to your customers. For every key aspect, there could be several or many expectations. Once you determine the essential expectations, figure out ways to measure how well your product or service is meeting those expectations. Then, work backward through the process, to establish metrics for activities that are critical to meeting those expectations.
Make sure that all measurements are linked to bottom-line results. For example, if you’re working on making your super widget more durable, you should focus on metrics for raw materials and assembly, but not on metrics for painting or packaging. For every metric, ask this question: How does this metric link to the bottom line?
With Six Sigma, this is a continual question—because Six Sigma is based on tangible financial results. Measurements must result in an identifiable impact on the bottom line.
Identifying CTQ factors is generally a laborious process. In one company, the Six Sigma team analyzed 800-1,000 processes, each of which had 100-120 specifications. It identified the most critical factors that would lead to greater customer satisfaction, lower costs, and/or greater ease of assembly. The team then mapped and prioritized CTQ factors to be targeted in Six Sigma projects.
Process Mapping

Mapping processes is extremely valuable. A process map is an illustrated description of how a process works, a flowchart of the steps in a process— operations, decision points, delays, movements, handoffs, rework loops, and controls or inspections.
It takes a lot of time and effort to identify CTQ factors, but without mapping each and every CTQ process, you won’t get the information necessary to target areas for improvement and fix problems.
Process maps allow a team to visualize the flow of products or the sequencing of activities. In so doing, it can locate steps that don’t add value to the process. Eliminating such steps is an easy way to reduce cycle time and cut costs.

Value-Added versus Non-Value-Added

A key step in making improvements that matter to the customers is to determine which processes add value and which do not—from the customers’ perspective. The concept of distinguishing between value-added activities and non-value-added activities is simple: any part of a process for which the customer is willing to pay is value-added, while any part for which the customer is not willing to pay—such as moving or storing raw materials or approvals by managers between processes—is non-value-added. Because non-value-added activities do little or nothing to satisfy customers and only add costs, they should be targets for elimination. In addition, whenever an activity can be eliminated, complexity is reduced. That means lower costs and fewer opportunities for errors.
However, it can be difficult to work with that distinction between value-added activities and non-value-added activities, particularly when the processes have been in place for a while and/or the people involved are secretive, defensive, and/or territorial. Nonetheless, Six Sigma enables you to show how much time and money the organization can save through eliminating non-value-added activities.
Are there any non-value-added activities in your organization? Bet on it! It’s been said that all processes accrete additional activities and complications that add little or no value, just as the hull of a ship accumulates barnacles, which make it less efficient. So, think of this part of the Six Sigma process as scraping the hulls of your business processes to gain efficiencies relatively easily.

Waste

Waste is essentially anything that customers don’t want to pay for when they buy your products or services. Here are some examples.
Activity. Sometimes, even often, activity is work only because the employees are being paid for it. The activity is adding no value to your products or services. Maybe it’s because a process is inefficient. Maybe it’s because a step is unnecessary. For each of your processes, think about what it’s intended to do, about the inputs and the required outputs, and then imagine designing a process to turn those inputs into those outputs. That zero-based thinking can help you identify steps in your current process that are adding no value, steps that are waste.
Waiting. When people or products or documents must spend time doing nothing between value-added steps, that’s waste. There’s waste in many processes—but that doesn’t mean that it must be tolerated, or that it cannot be eliminated.
Excess. Sometimes this type of waste is overproduction—making more of something than required by the next process or making it sooner than required by the next process. It’s waste when it takes people or equipment or materials away from more immediate tasks or when it necessitates inspection, storage, and/or transportation. In some cases, the product might not even be necessary and must be sold at a discount or scrapped. Occasionally, excess products get damaged or go missing. Sometimes excess is a problem when employees order more materials or supplies than are necessary within a given period—especially when the money used for those materials or supplies is needed for other purchases or when the need for those materials or supplies ends before the inventory is exhausted.
Transportation. Whether you’ve got forklifts moving widgets or inter-office mail moving documents, extra steps that do not add any value to your products or services are waste. Don’t work at improving the transportation; work at minimizing it.
Under-utilization of people. This can be a big waste—and it can be easy to neglect because the cost is usually in soft dollars—opportunities lost—rather than in hard dollars. Who could be doing things that would benefit the organization more and possibly allow the employees to feel more satisfied and improve morale? What abilities is the organization not tapping and developing? If conventional thinking, restrictive politics and culture, poor personnel policies and practices, little or no developmental training, or other factors are keeping your organization from taking advantage of the experiences and skills and enthusiasm of any employees, it could be missing out a lot.
These are the main types of waste in most organizations. When you map your processes and examine every activity in terms of adding value to your products and services from the perspective of your customers, you can probably identify a lot of waste. And, once you identify waste, you can work toward eliminating it

[sk_kireeti]

LESS IS MORE

When setting metrics, keep the numbers of measurements small. As you get into the Six Sigma mind-set, it’s natural to want to measure everything. Don’t! The key here is quality over quantity. Select only a true set of indicators that will give you the information you need about process factors that affect customer satisfaction and revenue. When establishing a metric, you need to know why you’re measuring it, why it’s important, and what’s causing the results.


Think back to the dashboard analogy. Driving would be more difficult if the dashboard contained too many indicators. The same is true with too many metrics. Select and limit measurements carefully. The essential indicators will provide the information you need.



Your metrics should provide data that enables you to identify and solve performance problems in your processes as quickly as practical. They should also, consequently, be sufficiently sensitive to reveal changes or variations that are significant.


What’s “significant?” That depends on your baselines and your goals. It’s important that your metrics capture change with enough sensitivity to enable you to take action.


For a simple example, let’s take our widgets. If the cycle time for the process of molding components averages three hours and seven minutes, a metric that tracks minutes might be sensitive enough. If the cycle time for the process of assembling those components averages 11 minutes and 54 seconds, then we’d want a metric that tracks seconds. If the cycle time for the process of packaging widgets averages four seconds, then our metric should be sensitive to at least tenths of a second.


Metrics need to “slice and dice” in increments that capture small but significant changes and track them in terms of cost, time, and quality. They indicate the ability of a process to achieve certain results—your process capability index, as explained earlier.

[sk_kireeti]

UNIVERSAL STANDARD

One of the innovations of Six Sigma is to establish metrics in terms of opportunities for defects. That is a metric that managers can use to apply Six Sigma thinking to any business process, because “defects” can be anything that causes customer dissatisfaction and “opportunities” can be any possibility for a defect. By calculating quality levels according to the complexity of the product, service, or process, Six Sigma allows for metrics that make it easier and more realistic to compare performances of products, services, or processes that differ.
Here’s a simple example. William and Mary both work for Acme Wax Fruit Company. William runs the apple production line, which melts wax cubes, pours the wax into molds, and then dips the resulting item into a wax bath of another color. Mary manages the shipping department for the citrus fruit division; she’s responsible for the employees who handle the inventory (oranges, lemons, and limes) in the warehouse, the employees who load the trucks, and the truckers who deliver the goods. The processes that William and Mary manage vary greatly, but because some of the metrics established for them are in terms of DPMO, it’s possible to know that they are currently at 81,900 DPMO and 74,700 DPMO, respectively, and set a goal for next year of three sigma— 66,8000.
And that’s how, with metrics using DPMO, you can compare apples and oranges!



BENCHMARKING

You’ve determined the value content of your processes, you know which affect CTQ customer issues, and you’ve established your baselines. Now you can move on to understanding how your processes measure up and figuring out where you want to go with your processes. You do that through benchmarking, both internal and external. Benchmarking is basically a method for comparing a process, using standard or best practices as a basis, and then identifying ways to improve the process. Through benchmarking, you can establish priorities and targets for improving the process and identify ways to do so.
At this point, you know what processes you want to benchmark and you’ve got your metrics. Now what?
The next step is generally to identify benchmarks for your target processes and to gather benchmark data to show how your processes could be performing. The benchmarks might be internal or external. Because most Six Sigma initiatives use benchmarks outside the company and because that’s a more complicated practice, external benchmarking will be our focus here. (If you decide to use benchmarks within your company, it’s considerably less difficult.) Whether you benchmark internally, externally, or both, it’s essential to ask key questions about why your performance differs and to determine how you measure defects and yield rates.
How do you identify benchmarks? First, you consider your competitors. Which among them has target processes that perform better than yours? You might know that through competitive intelligence or through media reports. Also, as you probably realize, the Web has become a great source of information on companies. Articles that would have gone unnoticed in small, local publications or data that would have been buried in a report are now out there for you to access.
After you’ve identified the benchmarks to use, you collect data on the target processes. How you do this depends on the processes, the benchmarks identified, and the sources of information. Use your creativity and investigative instincts and skills.
Think Creatively

Choose your benchmarks carefully. Don’t benchmark only very similar processes in the same industry unless you want to quickly improve a very defective process. Also, be cautious about benchmarking your competitors: that makes sense only if you know that their processes are better.
Think in terms not of the process, but of the purpose for the process. That way, you can identify dissimilar processes from which you can learn.
Here’s an example. An airline has problems with its baggage-handling process. It could benchmark other airlines, of course. But the project team could also think beyond baggage and study national package delivery services. Another good example for benchmarking is a company that was preparing to move its corporate manufacturing to another location across the country. The managers asked the question, “Who’s the best in the world at moving?” They selected rock band “roadies” as the benchmark for excellence in moving a complex setup efficiently from city to city! They studied the practices of this “best of the best” and adapted them to their particular situation.
To briefly show how benchmarking works, let’s consider a bank, specifically the residential loan processing department. Imagine that you’re a customer waiting for the department to process your application. You’re in a hurry. In other words, your CTQ requirement is promptness. But one week passes, then two, and then three …
The lending institution, by not meeting your CTQ requirement, risks alienating you (and probably many other customers) and increases its cost of services sold (COSS). Of course, if it raises its fees to cover those costs, it’s likely to lose even more customers.
Each functional group in the organization—customer service, sales and marketing, finance, information technology, accounting—plays a part in delivering what the customer wants. Each group has processes that should work together to serve the customer. Are those processes the best they can be? Clearly not, because loans are delayed by nearly a month. So the residential loan department needs to benchmark its processes against processes in other divisions to discover waste, so it can work to eliminate it.
Greg, the manager of the residential loan department, decides to start benchmarking internally. He determines which loan department is processing the most loans with the lowest defects. Then it’s a question of studying how the processes in that department work better and finding ways to improve his department’s processes.
Greg could also examine how the loan process works in competing companies. Maybe one aspect is better with Alpha Savings and Loan, another with Big Money Loans, and yet another with Consumer Loans.
You might be able to get information for benchmarking from public domain sources, through the library or the Web. Some companies provide information in white papers, technical journals, conference presentations, panel discussions, and so forth, or in materials for vendors and customers or advertisements. There is a lot of information available about what you are trying to benchmark from public domain sources. These sources include the following:

Library databases	Inquiry service
Internal publications	Literature searches
Industry publications	Internal reviews
Functional trade publications	Professional associations
Industry data firms	Special industry reports
University sources	Seminars
Newspapers	Industry experts
Newsletters	Company “watches”
Customer feedback	Advertisements
Original research	Networks
Telephone surveys	Web site searches

Internal publications Literature searches Industry publications Internal reviews Functional trade publications Professional associations Industry data firms Special industry reports University sources Seminars Newspapers Industry experts Newsletters Company “watches” Customer feedback Advertisements Original research Networks Telephone surveys Web site searches
An excellent organization that can help with your benchmarking exercise is the American Productivity and Quality Center (APQC). You can reach them at www.apqc.org or by phone at (713) 685-4670.
You might need to develop questions for a survey to be conducted by mail, by telephone, by fax, or by e-mail. You might decide to take the most direct approach, to contact companies and arrange site visits. You could also enter into a benchmarking partnership, in which each partner would gain information about the others in exchange for sharing information on its own processes. Another possibility is to work with a competitive intelligence firm.

Keep It Legal and Ethical

Benchmarking can be risky business. To minimize the likelihood of misunderstandings, ethical slips, and legal problems, you should follow the simple Code of Conduct scripted by the International Benchmarking Clearinghouse, a service of the American Productivity & Quality Center (www.orau.gov/pbm/ pbmhandbook/apqc.pdf). It provides guidance through outlining the following eight principles:

• Legality
• Exchange
• Confidentiality
• Use
• Contact
• Preparation
• Completion
• Understanding and action

Once you’ve completed your benchmarking studies, you should have data for each of your key metrics for the targeted processes. Then, you’re ready for the next step.
To sum up the benchmarking process, here are the basics:

• Select a topic for which to establish a benchmark.
• Estimate the costs of doing a benchmarking study.
• Select and train a team to complete the study.
• Choose the key metrics you want to study.
• Select or design a survey and develop questions.
• Develop tools to collect the data.
• Test the methods you plan to use to analyze the data.
• Analyze your own process for the key metrics.
• Identify companies and contact them.
• Collect and analyze data from public domain sources.
• Formulate a benchmark plan and analyze information.
• Conduct gap analysis.
• Develop recommendations and plan to implement.
• Identify departments that are needed for support.
• Present the plans to management.
• Implement the plan and monitor the progress.
• Collect data on benchmark subjects.

GAP ANALYSIS

You’ve used your key metrics to establish baselines for your target processes. You’ve gathered benchmark data to show how your processes could be performing. Now, you compare. In technical terms, you do a gap analysis, to quantify the gaps between where you are now and where you want to be.
Your gap analysis enables you to set goals for improving your processes and to develop strategies for improvement. You might not be able to set goals for improving every aspect of every process; you might need to prioritize. You might not be able to set ideal goals; especially early in Six Sigma implementation, you might want to set goals that allow you to achieve important gains quickly, to prove the value of Six Sigma. Then you can set your sights on breakthrough goals.