NOTE: TQM = "total quality management"

Most failures in TQM programs result from responsible (and otherwise thorough) people doing what seems simplest, or most obvious, or least risky to do. Measurements that are easy to take, results that are readily visible and understandable, actions that require minimal training -- these are common features of actions that lead to the collapse of TQM programs:

• measuring the wrong quantities (those easily observed and visible to all)
• gauging behavior instead of results
• emphasizing courtesy instead of competence (easy to achieve)
• attaching cost-cutting riders to the program
• failing to understand and trace cause and effect
• failing to benchmark
• following inconsistent strategy
• depending on performance appraisals to determine individual contributions
• offering bad incentives (e.g. equivalent raises and bonuses across the board)
• using slogans and hype
• appointing a "Chief Quality Officer"
• instituting a separate implementation organization
• giving people bad jobs
• depending on overspecialization

Results which can and should be measured fall into several categories (not just profit and/or cost):

• customer satisfaction
• operational efficiency
• product quality (based on measured characteristics)
• finances
• employee satisfaction
• public responsibility (which may determine government cooperation and/or contracts)

Measurements can sometimes doom (or at least burden) quality initatives as well:

• from too much or too little data
• from decisions made apart from data or inconsistent with data
• "unspoken measurements" (e.g. whether a worker is a "team player")

References

The sci.engr.* FAQ on Engineers and Quality
Brown, Hitchcock and Willard. Why TQM Fails. Burr Ridge, IL: Irwin Professional Publishing, 1994. ISBN 0-786-30140-6