Key #4 BaseDecisions on Data and Facts

Overview

There are a lot of good reasons why data and facts form the true foundation of Lean Six Sigma. Want to know who your customers are and what they want? You need to collect data. Want to improve processes? You’ll need to collect data on variation, defects, and process flow. Want to avoid the kind of needless arguments and squabbling that destroy teamwork? Have a rule that people must support their opinions with facts.

You also need data and facts because they’ll save you a lot of trouble and prevent a lot of wasted dollars and time…

When the utilities in one state were de-regulated, one company suffered a great deal of “churn,” losing customers about as fast as it gained new ones. They were forced to spend a lot more money on marketing now that they had to compete for customers.

The customer service staff had noticed several cases where new customers came on board then changed their minds right away, ultimately switching to a different company. Very quickly, these examples grew into a widely held assumption that new customer turnover was the reason behind the churn. The new customers were, the reasoning went, targets of marketing efforts by rival power suppliers.

The service staff therefore began focusing on how to keep these new customers from switching. They developed a new Welcome Pack explaining their services and benefits, which they began sending to thousands of new customers every week. At a cost of $8 a piece, this packet represented a significant investment.

At one point, however, a Lean Six Sigma team at this utility company collected data on churn. They found that new customers only accounted for about 4% of the total. The other 96% were long-term customers who were switching utility companies. In other words, the company was spending thousands of dollars each week on something that would solve only 4% of the problem! So they re-directed their marketing efforts to try to keep the customers they’d had for some time.

This company’s experience is common. Their initial decision about what to do was wrong because they made it without data. Having data can make a huge difference in the decisions we make every day on the job, and is particularly important in improvement projects. Unfortunately, learning the data habit is harder than it has to be because of a number of roadblocks:

1. A lack of available data. Do you know how much work you have in your process at this very moment? Do you know how long, on average, it takes you to handle those work items, be they phone calls, requests, reports, bills, orders, etc.? Do you have a way to find out? Do you know what your work group’s error rate is on average? How many of your customers are happy with the product or service you provide? How many are unsatisfied? Could anyone in your office answer questions about the quantity, quality, and speed of their work? For most people, the answers to those questions are “no” or “I don’t know.” People working on early Lean Six Sigma projects are often starting from scratch when they begin to collect data.
2. Little training in collecting or analyzing data. If you’ve never had to collect data before, the number of decisions involved will probably surprise you. What data should you collect? How can you make sure it will answer the question your team is asking? How should you measure what it is you’re trying to measure? How often should you take a measurement? How can you make sure that everyone collecting data will do it in the same way? Once you have the data, how can you analyze it to understand what it’s telling you? There’s a lot to learn!
3. A historical pattern of using data only to punish or reward individuals, not to make decisions about improvement. For many years, a lot of organizations have used data for only one purpose: to punish or reward people. Did you meet the sales quota? No? Then your pay will suffer. Did you finish all your customer calls in 60 seconds or less? Yes? Then you’re a star. What’s very different in Lean Six Sigma organizations is that data is used for learning and for monitoring process performance. You will be collecting data to understand what’s going on in the process, where problems are arising, and what solutions will really work. Once improvements are made, you’ll also be collecting data to track how well the process is doing, to detect any early signs of trouble, and to help you maintain the gains you’ve already made.

What kinds of data?

Once your company has made the commitment to collect data, the obvious question is “what kind of data?” Making that call is something you’ll learn about if you go through training or participate on a team. To jump start your own thinking, we’ve given examples of actual data collected by teams in Chapters 7 and 8. In general, it all falls into two categories: result measures and process measures:

• Result measures reflect the outcome of a process or procedures—how the product or service turned out
• Process measures reflect what goes on to produce the result

In a baseball game, for example, the final score is a “result” measure. Stats like hits, errors, strikes, and walks are all “process” measures. They tell you what went on during the game to produce the final score.

You need both results and process measures to be effective in Lean Six Sigma. You absolutely have to keep track of the final result. But the only way you can improve a result is to change the process, and you’ll need process measures to tell you what has to change and how.

What should you actually measure? Here are four typical types of data that teams find useful:

1. Customer satisfaction (a result measure): Data gathered through surveys or interviews on what customers think about your product or service, and what it’s like doing business with your group or organization.
2. Financial outcomes (a result measure): What impact the quality and/or problems have on revenue, expenses, costs, etc.
3. Speed/ lead time (result or process measure): Data on how fast (or slow) your process is. “Lead time” is how long it takes for any individual work item to make it all the way from the beginning to the end of the process (when it is delivered to the customer). If measured at the end of the process, speed is a result measure. If measured on individual steps, it becomes a process measure. (You’ll read more about speed and lead time in the next chapter.)
4. Quality/defects (result or process measure): How many errors are made, whether the product or service has flaws that affect the customer, and so on. Like speed, quality can be a result measure if the data are collected on the final product or service. But most teams also use it as a process measure, collecting data on what happens within the process.

Won’t Gathering Data Slow Us Down?

At the end of its project, one team working on a purchasing problem realized that it had spent 75% of its project struggling to get good, reliable data. When some people hear a number like that, their first reaction is “we can’t afford to spend that kind of time just gathering data!”

That kind of reaction is short sighted. It was BECAUSE of the time they invested in getting good data that the team in question could solve a problem that had been around for years. Getting the right data also allowed the rest of the project to go quickly. Whenever the team faced a decision such as “what solution should we try?” they could look at the data. So discussing the data replaced the kind of endless arguing that happens in teams who don’t use data!

Skipping the data collection step is NOT an option in organizations that are really serious about Lean Six Sigma.

Conclusion

Roger Hirt, a Six Sigma specialist who works with the City of Fort Wayne, Indiana, was sitting in on a city panel meeting once where a city employee was reporting on an ongoing project. During the meeting, an influential member of the panel piped up to offer a solution. The employee thought about the suggestion and said, “I guess that would be possible.” But then Roger stepped in. “Just a minute. We have to look at what the data tells us about the problem before we’d know whether that solution would do any good.”

It’s impossible to go back through history, or even look at organizations today, and see how many bad decisions were made because people didn’t gather data. The number would be astronomical. Today, organizations that are using their resources most effectively insist on using data as often as they can.

But it’s a hard habit to learn because we’re so used to not collecting data. We have to re-train ourselves to pause before making a decision and think about whether there is existing data we could look at, or if we need to collect new data. Learning to ask one simple question—“What does the data tell us?”—will make a huge difference in your improvement efforts.