Wednesday, January 30, 2008

The 10 Biggest Quality Mistakes




One of the most effective ways to improve is to learn from other people’s mistakes. Experience has taught me that there are plenty of mistakes out there to learn from. The trick is to recognize them and understand what to do instead. Unfortunately, I keep seeing the same mistakes over and over. They aren’t mistakes because they violate a standard such as ISO 9001; they’re mistakes because they violate good sense. Let’s examine the 10 most common quality mistakes and see how they can be corrected.

1. Limiting quality objectives to traditional quality topics

The term “quality objective” is an unfortunate one. It introduces subjectivity (i.e., quality) into a subject that should be quite clear (objectives). A much better term would simply be “measurable objectives” because that requires less interpretation. The word “quality” clouds the issue and makes many people want to narrow the focus of what a quality objective can be.

The truth is that quality is reflected in everything an organization does, and a quality objective can be anything measurable that relates to the organization’s success. A quality objective might relate to finances, customer feedback, safety, efficiency, speed or innovation. All these attributes relate to quality in one way or another. When selecting quality objectives, organizations should examine what matters most to their success. Whether the resulting measure is tied to traditional quality control or quality assurance is irrelevant.

2. Holding infrequent management reviews

Management review is the process used by top management to analyze data, make decisions and take action. Ideally, it’s a preventive process because data should indicate threats before they blossom into full-blown problems. If top managers are unable to analyze data proactively and prevent problems, then they’re not doing their jobs. Holding management reviews once or twice a year ensures that actions taken won’t be preventive. Only through timely and frequent data review can actions be preventive. Once or twice a year won’t cut it.
Many people argue that their organizations already review data on a weekly and monthly basis. This means that management review is still after the fact; the decisions have already been made. Management review must be included in existing meetings. Instead of a twice-yearly dog-and-pony show, cover the inputs and outputs of management review as they occur naturally during existing meetings. After a month or two, you’ll have addressed all the required inputs and outputs. Using this approach, you’ll have information that’s timely and resulting actions that are preventive. Another advantage is that you dispense with the need for long, laborious management reviews. They happen in a smooth and effective manner that’s much more likely to drive improvements.

3. Sending out long, complex customer surveys

The days of the long and complicated customer survey are over. People don’t have time to complete them. Even when organizations design shorter surveys, the questions are often confusing and fraught with interpretation problems. The scales that accompany the questions are often unbalanced and illogical. As a result, organizations end up with a small amount of valid data. Better to have no data at all than data that could lead you in the wrong direction.

Instead of a survey, why not simply ask your customers what they like and dislike? Don’t limit their responses to survey topics. Let your customers dictate the content of their feedback in response to open-ended questions. Few are more powerful than the following: What do you like? What don’t you like? What would you like to see different in the future? Open-ended feedback is also much easier to understand and take action on. A customer- satisfaction index of 3.8 is hard to interpret. On the other hand, seven out of 10 customers telling you that your Web site is confusing is very easy to interpret.

4. Assuming everyone knows what “nonconforming” looks like

When I visit organizations, one of my favorite questions is, “Where do you put the nonconforming products?” Control of nonconforming products is one of the most basic kinds of controls, and it speaks volumes about the rest of the controls embraced by the organization. Unfortunately, where I often find nonconforming products is wherever someone decided to leave them. They’re not uniquely identified, either. In other words, nonconforming products are treated no differently than any other products. When I ask why this is happening, the most common answer I get is, “Because everyone knows that’s nonconforming.” No, they don’t. No matter how nonconforming a product is, there will be someone who won’t recognize it as such. The stories of bent, broken and blown-up products that somehow got used anyway aren’t urban legends. They’re true.

Smart organizations positively identify all nonconforming products, and really smart organizations segregate them to remove all chance of accidental use. Error-proof your control of the nonconforming product process so that nobody has to assume anything.

5. Failing to use the corrective action process

Corrective action is the systematic way to investigate problems, identify their causes and keep the problems from happening again. Nobody wants problems, but it’s essential to have a way of dealing with them when they come up. The more the corrective action process is used, the better the organization gets at addressing its risks and problems. That’s why I’m astounded when I hear about organizations that avoid using their corrective action processes. Of course, I always ask why they’re doing this, and I often get one of these answers:
• Corrective action isn’t effective for large problems.
• Corrective action isn’t effective for small problems.
• Nobody understands root cause.
• Our problem-solving tools are confusing.
• Our procedure requires too much paperwork.
• Corrective action takes too long.
• I hate our corrective action form.
• Top management frowns on corrective action because it means that someone screwed up.

These aren’t corrective action problems but problems with the organization’s approach to corrective action. An effective corrective action process is typically seamless, simple and intuitive. The whole point is to add a little structure to problem solving, not to create additional bureaucracy.

Here are some hints to make your corrective action process more user friendly and effective:
• Strip it down to the essentials. A corrective action must clearly describe the problem, how it’s caused, actions taken to remove the causes, results of actions taken and how the actions were effective. Only include additional elements when you can prove that they add value.
• Remove all jargon from the process. Strange words only discourage people from using the process.
• Don’t insist on a raft of signatures . It’s not necessary for half the management staff to sign off on every corrective action.
• Remove paper from the process as much as possible. Use electronic media to track corrective actions.
• Communicate corrective actions widely. When people see that corrective actions accomplish something, they’re much more likely to participate in the process.
• Provide problem-solving tools, but give people some discretion in their use . If your procedure requires a failure mode and effects analysis to be completed for every corrective action, it will probably discourage people from starting corrective actions.
• Use teams for corrective actions whenever possible. This gives people experience in the process and also increases the effectiveness of most solutions.

6. Applying document control only to official documents

Most organizations do a decent job of controlling “official” documents, the procedures and work instructions that form the core of the quality management system (QMS). These are often written, approved and issued according to very specific guidelines. What organizations don’t do very well is control unofficial documents, many of which are more important than the official ones. What am I talking about? Here are some examples that are often found in production areas:
• Post-it notes with special requirements written on them
• Memos that include procedural steps
• E-mails with customer specifications
• Photographs showing what a product should look like
• Drawings indicating how product components fit together
• Samples of product showing defect limits

These informal resources become documents when they’re shared for people to
use, and they’re some of the most important documents within an organization. They’re distributed and posted in a hurry--usually without control--because the information they communicate is considered critical. Nobody can quibble with the speed of distribution, but the lack of control guarantees problems later. I’ve seen 10-year-old memos posted in organizations that exhorted personnel to perform obsolete job steps. When documents aren’t controlled, mistakes and nonconformities are inevitable. Apply document control to all documents, and scrutinize your document control process to keep it streamlined and effective.

7. Focusing audits on petty, nonstrategic details

Auditing is the process of comparing actual operations against commitments the organization has made. It’s a simple, fact-driven process that can generate huge improvements. However, these can occur only if auditors focus on the right things. Too often, internal auditors become preoccupied with petty details and neglect the big issues. They’re uncomfortable examining the big, strategic issues. It’s much easier just to nitpick. Organizations rarely provide enough training and skill-building to their internal auditors, so it’s no wonder that they aren’t prepared to carry out their duties to the fullest.

A robust internal auditing process examines make-or-break issues. Here are just a few of the items that internal auditors should probe in detail:
• Customer satisfaction. Is the organization capturing and analyzing customer satisfaction data? How is it acting on the data? Do trends show improvements in customer satisfaction?
• Management review. Does management review happen as planned? Does the necessary information and data get reviewed? What actions result?
• Corrective action. Is corrective action applied to existing nonconformities? Is it timely? Does evidence indicate that causes are removed to prevent recurrence?
• Preventive action. Does the organization take preventive action based on data and information? Is it effective?
• Internal audits. Are audits scheduled and carried out based on status, importance and prior audit results? Do audit nonconformities become corrective actions? Is the entire scope of the management system audited?
• Objectives. Are objectives established and communicated? Do employees understand them?
• Control of nonconforming products. Are all nonconformities positively identified? Are dispositions carried out as required? Are trends analyzed for improvement?

There are, of course, many other important issues an audit process could examine. The point is that internal auditors should go after the items that really affect the organization’s success. Focusing on petty details serves no purpose but to confuse everyone about the purpose of audits.

8. Training some personnel, but not all

Most organizations provide significant training to hourly production personnel. Salaried and managerial personnel are often neglected, however. Why? Because there’s a perception that salaried workers don’t affect product conformity. This is a serious error.

All personnel must be included in the training process. Salaried and managerial personnel need more--not less--training because their decisions and actions have more lasting effects. When an hourly employee makes a mistake, it could cost money. When a top manager makes a mistake, it could put you out of business. Doesn’t it make sense to train these people? Do it early and often.

9. Doing anything just because an external auditor told you to

External auditors wield great influence. Their statements and judgments can have a lasting effect on the way an organization conducts business. This can be good or bad. Usually, it’s bad. Most external auditors working for a registrar are removed from the realities of running a business. They travel from organization to organization, gradually collecting paradigms about the way a QMS should be implemented, maintained and improved. These paradigms are sometimes reflected back to the organization in the form of recommendations or nonconformities.

In my travels to companies, I often ask people why they’re carrying out a process the way they are. I always raise this question when the process seems unwieldy or illogical. In a surprising number of cases, the answer will be, “Because the external auditor said we should do it that way.” What a waste. Do a reality check on the auditor’s recommendations. Never do anything just because an auditor would like it done that way. A certificate on the wall isn’t worth it.

10. Employing someone who only oversees the QMS

Having a person who does nothing but oversee the ISO 9001 (or any other) QMS is one of the worst ideas in the history of quality. Why? Because it guarantees two things:
First, the QMS coordinator will become isolated from the rest of the organization. Because the person does nothing but serve the QMS, he or she loses touch with why the organization exists in the first place. The system becomes paramount over the organization’s business concerns. Second, the QMS will become bloated and bureaucratic because it must expand to completely fill someone’s time. Procedures become more complicated, methods more cumbersome and the benefits more ambiguous.

A QMS is nothing more than a guiding structure of methods, and it shouldn’t take a huge dedication of time and effort to maintain. Yes, someone should keep the system on track, but that person should have other responsibilities as well. Pair the ISO 9001 coordinator job with other responsibilities that focus on understanding what the organization does, especially responsibilities related to the product, customers and improvement. If the QMS is so bureaucratic that it requires the time of an entire person (or, heaven forbid, an entire staff), then the system needs to be streamlined. An effective QMS should make an organization more competitive, not weigh it down.

4 comments:

jelly1921 said...

I personally made a translation of the article from English to Chinese without author's permission, which could be found @ http://jelly1921.blogspot.com/2009/02/10-biggest-quality-mistakes.html

jelly1921 said...

I post the article into Chinese quantliy forum and received lots of appraises. It would be treated as a part of training content, I was told.

Thank Craig

tnerB said...

Is this a book as well?

James jones said...

Hello! many thanks for the nice writing vogue. Keep it up! medical device supplier quality