Risks, Possibilities, and Best Practices
Steve Kerr Provost Chancellor University
Douglas LePelley Associate Professor of Business Nyack College; Senior Director and Dean of Graduate Studies, Chancellor University
In our roles as leaders, managers, and decision makers, we are all confronted with numerous choices. Among the most important of these pertains to the nature of the goals we establish. Specifically, whether setting goals for ourselves or for others, we may elect to set small, continuous-improvement goals; realistic but challenging goals; or goals that are so daunting that no methodology for achieving them currently exists.
Impossible only defines the degree of difficulty.
—Herb Brooks, coach of the 1980 US Olympic hockey team that upset a heavily favored Soviet team to win the gold medal
I won’t let you down. There’s no way you can.
— communication between Colonel Potter and Frank Burns, MASH TV show
A great deal of research has been conducted on the relative merits of assigned versus jointly determined goals, quantitative versus qualitative goals, short-term versus long-term goals, and other important components of the goal setting process. One of the most important choices to be made is the level of goal difficulty, which has been found in numerous studies to exert significant influence on employee attitudes and performance.
One of the most consistent findings concerning the level of goal difficulty is that when goals are set too low, people often achieve them, but subsequent motivation and energy levels typically flag, and the goals are usually not exceeded by very much. Compared to easy goals, difficult goals are far more likely to generate sustained enthusiasm and higher levels of performance. However, this finding comes with an important caveat, namely, that the goals, though difficult, must be seen to be achievable by those who are supposed to attain them. The logic behind this finding is nicely expressed by Locke and Latham, who point out that
Nothing breeds succeeds like success. Conversely, nothing causes feelings of despair like perpetual failure. A primary purpose of goal setting is to increase the motivation level of the individual. But goal setting can have precisely the opposite effect if it produces a yardstick that constantly makes the individual feel inadequate. Consequently, the supervisor must be on the lookout for unrealistic goals and be prepared to change them when necessary.
( Locke & Latham, 1984, p. 39)
Consistent with Locke and Latham’s observation, considerable research shows that overly difficult goals can cause people to view them as ridiculous and not take them seriously, or alternatively, to work harder and harder until they eventually conclude that the goals cannot be attained, whereupon they become demoralized and disengage (c.f. R einertsen, 2000; Choo, 2011).
Yet, in the face of such definitive research, and against all common sense, some organizations persist in intentionally constructing goals whose performance levels are so high that the great majority of attempts to achieve them invariably end in failure. These goals, called “stretch goals,” will be the primary focus of this chapter.
The notion that stretch goals have the potential to increase an organization’s aspirations and subsequent performance has been around for a long time. In one of the most influential business books of the 1960s, Cyert and March noted that most organizational goals are influenced primarily by three factors: the organization’s past goal, the organization’s past performance, and the past performance of other “comparable” organizations ( Cyert & March, 1963, p. 115). Obviously, conceiving goals in such a manner is unlikely to stimulate innovation or result in breakthrough ideas. The authors pointed out that by establishing goals that are less tied to the past, and by recognizing that there is an interaction between the level of the goals that are set and resultant employee behavior, “an organization can induce behavior designed to confirm its prediction (goal)” ( Cyert & March, 1963, p. 112).
In their discussion of the implications of Cyert and March’s work, Sitkin et al. noted that “by forcing a substantial elevation in collective aspirations, stretch goals can shift attention to possible new futures and perhaps spark increased energy. They thus can prompt exploratory learning through experimentation, innovation, broad research, or playfulness as organizational actors seek new or varied approaches to reach the target” ( Sitkin, See, Miller, Lawless, & Carton, 2011, p. 545). However, on the basis of their review, Sitkin et al. asserted that the business literature has overweighted the successes attributed to stretch goals and underreported the cases where setting stretch goals turned out to have no effect or were counterproductive.
The concern expressed by Locke and Latham—that employees can easily become demoralized if they believe the goals set for them are unachievable—is especially relevant to the concept of stretch goals, because such goals are not accidental overreaches that the supervisor “must be on the lookout for,” but are intentionally set at levels that are “seemingly unattainable with present resources, … and not currently seen as possible” ( T&D Guru, 2011 ). Talk about potential frustration! At GE, Jack Welch used to tell his managers: “If you’re making most of your stretch goals, all that tells me is that you aren’t setting them high enough” ( Welch, 1990–1991 ).
Why, then, if stretch goals have such high potential to create dysfunctional outcomes, did some of the supposedly best-managed, most people-oriented organizations in the1990s persist in using them? As a partial response to this question, let’s consider just a few of the experiences and reported results from companies that made use of stretch goals during this period—starting with GE.
Despite the attention paid to the topic by Cyert and March and others, the concept of stretch goals did not attract widespread attention until its adoption by GE’s then CEO, Jack Welch. When Welch first assumed the chairmanship of GE in 1981, the company had a very traditional approach to the goal setting process and the goals that were set. In common with many other corporations at the time, GE taught its new managers to adhere to the “SMART” model when setting and evaluating their goals. As employed by GE, the acronym START was intended to serve as a reminder to the company’s managers to set goals that were Specific, Measurable, Attainable, Results-Oriented, and Timely, so that they could be achieved in a maximum of three months ( General Electric, 1984, pp. 132–133). As a way to ensure that goals were attainable, GE told its managers to use the following questions as guidelines: “Is the goal realistic? Is it feasible? Can it be completed within the time allowed with the available resources?” GE managers were also presented with an example of an “unattainable” goal—something that was not to be used in GE. The example they were given was “to have no absenteeism for one full year” (p. 132).
As a result of this advice, GE’s goals were primarily influenced by the same three elements that, according to Cyert and March (1963), so may other organizations depended on: the organization’s past goal, the organization’s past performance, and the past performance of other “comparable” organizations. Not surprisingly, setting goals in this manner failed to consistently generate the out-of-the-box thinking and breakthrough results that Welch felt were needed to attain or maintain GE’s position as #1 or #2 in every market the company served.
For this and other reasons, by the early 1990s Welch had become very interested in some of the concepts and techniques of Japanese-style management. In particular, GE began to make extensive use of employee empowerment (called “Work-Out” in GE), continuous improvement (operationalized in GE as “finding a better way every day”), Six Sigma, and what Welch referred to as the “bullet train” method of goal setting. To underscore the importance of “bullet train thinking” in his discussions with his managers, he would often cite the following example:
It used to take more than six hours to travel by train from Tokyo to Osaka. If the Japanese executives had said to their engineers “I want you to reduce the time to six hours,” the engineers would have instinctively thought in terms of small improvements, perhaps in the way they boarded passengers and unloaded baggage. But instead, the Japanese executives set out a challenge to reduce the time of the journey to three and a half hours. Faced with such an “impossible” goal, the engineers and designers were forced to reexamine the most fundamental assumptions governing rail travel in Japan. The result of this reexamination was the bullet train.
( Welch, 1990–1991 )
To Welch, stretch goals were the mechanism through which he could use bullet train thinking, with all its attendant creativity, longer-term focus, and sense of daring, to fundamentally alter GE’s business processes, thus permitting the company to better compete with its Asian and best American competitors. By using stretch goals as a supplement (Welch might say antidote) to GE’s traditional goal setting and budgetary processes, he hoped to expand the unconscious upper limits he believed his managers were placing on productivity gains, quality improvements, cycle time reductions, inventory turns, and customer satisfaction. The ultimate purpose of stretch goals, he said, was “to reinvent how we do work” ( Bartlett, 1999 ).
Well before the use of stretch goals became widespread in GE, Welch required that his most senior business leaders present updated results in a number of areas at each quarterly meeting of his Corporate Executive Council (CEC). These results were of three kinds:
As the use of stretch goals became increasingly prevalent in GE, they began to exert influence over all three types of metrics. It is worth emphasizing that they were not employed as substitutes for the “regular” goals that GE managers absolutely had to meet, or risk losing their jobs. Rather, the stretch goals were intended to catalyze creative thinking and stimulate unconventional solutions. It could be argued that, in Welch’s mind, GE’s traditional goal setting process would bring about high performance today, while the stretch goals were for the purpose of ensuring high performance tomorrow.
An early example of Welch’s use of stretch goals was to announce to his managers that the new yardstick for return on equity (ROE), which he was certain they could achieve, was 20%. When he made that statement, GE’s ROE had already risen to a remarkable 18.7%, compared to an industry norm of 13.2%. Although Welch himself might have been confident that such performance was achievable, to most of his managers, a 20% ROE in a company as highly diversified as GE seemed impossible ( The Economist, 2011 ).
Many of GE’s stretch goals pertained to increasing the quality of the company’s products and services, and were expressed and quantified in Six Sigma terminology ( Harry & Schroeder, 2005 ). Welch observed at the time that “it is the struggle to meet stretch goals that helped US companies become world-class competitors again” ( Litvan, 1997, p. A-1). GE managers rose to the challenge—surprising themselves in the pro-cess—and accomplished a high enough percentage of the established stretch goals so that, in Six Sigma quality initiatives alone, GE produced savings of $12 billion between 1996 and 2000 ( Litvan, 1997 ). According to The Economist (2011), what Jack Welch and GE leaders learned from these early experiences was not to put any boundaries on productivity gains, quality improvements, or increases in market share. They concluded that if the right environment was created for the group, setting stretch goals and working toward what might seem to be impossible results often became reality.
During this period, a number of other organizations were also experiencing positive results from their stretch goal efforts. For example, in 1988, Commonwealth Health Corporation (CHC) formed a partnership with GE to improve their organization in a similar manner to what Welch accomplished in GE. As part of the process, CHC introduced their employees to various Six Sigma techniques, and established stretch goals in each of the following categories: customer satisfaction, timeliness of care, speed of care, convenience, quality of care/service, and overall cost. Beyond realizing improvements in each of these areas, by employing stretch goals in combination with Six Sigma, a culture of seeking significant improvement became a way of life at CHC. According to the company’s calculations, $900,000 was invested in the first year of the stretch goal program and resulting Six Sigma projects. The first year payback was $3 million, which grew to $7 million after Year 2 ( Thomerson, 2002 ). In reviewing the results of the program, Commonwealth’s CEO John Desmarais noted: “It is amazing, the use of stretch goals and Six Sigma. It is just a different way to look at things” ( Thomerson, 2002, p. 297).
Another company that was an early adopter of stretch goals, and had great success in applying the stretch goals concept to their Six Sigma quality initiatives, was Toyota Motors. From 1997 to 2001, Toyota employed stretch goals to induce their employees to look deeply into the underlying reasons behind the company’s high supply chain, distribution, and inventory costs. Rather than setting specific improvement goals in these areas, Toyota leadership challenged their employees to “dream big.” By their own calculations, the company realized more than $100 million in cost savings in these areas within two years ( Oxnard, 2004 ).
For more recent examples of organizations that have enjoyed great success by employing stretch goals, see Simkin et al. (pp. 545–546). However, keep in mind Simkin et al.’s assertion that stretch goals are, in their view, equally likely to induce negative reactions on the part of employees.
Up to this point we have noted that goals that are viewed as impossibly difficult may be dismissed as naïve or absurd, and simply ignored. If, instead, a stretch goal is taken seriously, employees may expend so much time and energy trying to meet it that they burn out in the attempt. If they do stay the course, experiencing continual failures may generate reduced self-confidence, and cause employees to become alienated from the organization.
Consistent with this latter argument, some theorists and researchers have argued that the creation of stretch goals can be emotionally harmful to many employees. Recent neuroscience research shows that the brain works in a protective way, such that any goals that require substantial behavioral change or thinking-pattern change will automatically be resisted. The brain is wired to avoid pain or discomfort, including fear. Fear of failure initiates a desire to return to known, comfortable behavior and thought patterns. According to this line of reasoning, the best way to initiate changes that the human brain will support is via small and incremental improvements, accompanied by positive reinforcement ( Williams, 2010 ).
Researchers have also observed that the psychological manifestations of not achieving goals may well be more damaging than would result from not having any goals at all. When people fail to attain something that they want and work for, their brains’ nervous systems often trigger strong negative emotions. Since the goal setting metrics of many organizations effectively create an either-or polarity of success, whereby goals are viewed as either met (via 100% attainment) or missed, and since stretch goals are set at levels that make 100% attainment very unlikely, the process virtually ensures that organization members will experience a loss of self-confidence and other negative emotions ( Williams, 2011 ).
A relatively obvious problem with stretch goals is that they are difficult to reconcile with the pursuit of work–life balance. If stretch goals are only utilized occasionally, for specified, limited periods, this may not be a big issue. However, in organizations that employ them on an ongoing basis, employees may be required to make fundamental choices between their work and home lives. In a 1995 Fortune interview, Steve Kerr, who was running GE-Crotonvillle at a time when numerous organizations were studying and seeking to copy GE’s approach to stretch goals, observed that “it’s popular for companies today to ask their people to double sales or increase speed to market threefold. But then they don’t provide their people with the knowledge, tools, and means to meet such ambitious goals …. To meet stretch targets, people use the only resource that’s not constrained, which is their personal time” ( Sherman, 1995, p. 231).
In his book W inning, Jack Welch admitted to practicing “do as I say, not as I did” by coming out strongly in favor of work–life balance ( Welch, 2005, p. 313). However, Welch went on to say that “your boss’s top priority is competitiveness. Of course, he wants you to be happy, but only inasmuch as it helps the company win. In fact, if he is doing his job right, he is making your job so exciting that your personal life becomes a less compelling draw” (p. 316). Welch then described three “successful” work–life balancing acts during his days in GE, but in two cases the balance was achieved by the employee turning down big promotions, and in the third case the employee “balanced” his work and home lives by coming in very early every morning and not leaving until six pm. A skeptic might point out that these are not really examples of a balanced solution (except perhaps over a period of years), since in each case, the time and energy invested in the employee’s career or home life came at the expense of the other.
Perhaps the most articulate spokesperson for this line of argument is Harry Levinson, who has long argued, with some empirical support, that to be effective, goals should be (a) team versus individually focused and (b) set by the group and/or by the employee and the manager jointly, but never solely by the manager ( Levinson, 2003 ). With respect to the need for goals to be team focused, Levinson noted that:
Most job descriptions are limited to what employees do in their work …. The more employees’ effectiveness depends on what other people do, the less any one employee can be held responsible for the outcome of individual efforts …. The reason for having an organization is to achieve more together than each could alone. Why, then, emphasize and reward individual performance alone, based on static job descriptions?
( Levinson, 2003, pp. 78 and 91)
Levinson’s assertion that goals must be established with extensive employee involvement stems from his conviction that the most useful performance feedback comes from peers, in part because higher management will invariably be unaware of structural impediments and other significant barriers to achievement. Lacking such knowledge, attempts by one’s boss to establish goals, and to assess and provide feedback about goal performance, are likely to generate employee cynicism. One of the implications of Levinson’s ideas for stretch goals is that, since they are typically set by higher management with little or no input from those who have to achieve them, they are unlikely to generate feelings of employee ownership or buy-in.
Taken as a whole, the concerns about stretch goals articulated above are substantial, and are buttressed by theory, research, and common sense. They are, however, primarily criticisms of stretch goals as they are typically conceived and employed, rather than constituting inherent flaws that cannot be prevented or remedied. In the remainder of this chapter, we shall discuss a few ideas, already in use, for mitigating these concerns and making stretch goals more effective.
One of the most serious charges levied against stretch goals is that they are typically constructed within systems that measure employee performance in terms of goal attainment, but are set at levels that make goal attainment exceptionally difficult. Consequently, stretch goals tend to produce adverse physiological effects in people’s brains and nervous systems, resulting in a variety of negative psychological consequences, including feelings of inadequacy and loss of self-esteem.
The truth and severity of these charges is undeniable, but remedies are available, and were successfully applied in a number of organizations. In GE, for example, it became evident very soon after the introduction of stretch goals that the company’s usual (harsh) reaction to missed targets would be not only unfair, but unproductive. Consequently, a new set of metrics began to be taught to all GE managers, and became a part of the business reviews. These metrics were designed to replace the “goals met/goals missed” either-or mentality discussed earlier in this chapter. They were built around a mantra that was created as a guide to be employed when assessing performance against the company’s stretch goals. The mantra said: “Don’t punish failure.”
Instead of asking whether an employee’s stretch goals were made or missed, GE managers were asked to consider the following three questions:
However, there is a caveat that accompanies employment of the mantra “don’t punish failure.” It is this: The mantra is only feasible when the stretch goals you set are well above the level of performance you must achieve to satisfy one of your stakeholders. If, for example, a new government regulation requires you to make substantial alterations to your company’s financial reporting systems, or take immediate actions to improve product safety, “not punishing failure” is a luxury you can’t afford. In such cases, it is essential that goals be set that at least minimally meet your stakeholders’ requirements, and it may well be a good idea to let people know that in this case failure will, in fact, be punished—if not by you, then surely by the government. Since, by definition, any aspect of your business that is given special emphasis by your stakeholders is one in which you want to excel, it may make good sense for you to establish stretch goals in these areas as well, but these stretch goals cannot act as substitutes for goals that absolutely must be met.
The process of establishing goals, including stretch goals, has also been indicted for failing to recognize how interdependent most employees are upon one another, and for denying meaningful influence into the process by peers and subordinates. These flaws do exist, but are not inherent in the process.
By way of illustration, soon after initiating stretch goals in GE, it became apparent that decisions needed to be made regarding how one division or department should react when another department established stretch goals that affected both of them. For example, the regional offices in many GE businesses were induced to establish sales goals that went well beyond what had ever been done before. This posed a ticklish problem for Manufacturing. Should they reconfigure their plants and accumulate raw materials in anticipation of these sales materializing? If Manufacturing didn’t do this, wouldn’t they be inadvertently sabotaging whatever chance Sales had to meet their targets? On the other hand, if Manufacturing did increase production and the hoped-for sales didn’t materialize, shouldn’t the added manufacturing costs be transferred to someone else’s budget—to Sales, perhaps, for failing to meet their goals, or to GE’s executive office, which had caused the $#%#!! stretch goals to be set in the first place? And if the added manufacturing costs (and for that matter, the added costs incurred by the regional sales offices because of the stretch goals) were not transferred to the executive office, wouldn’t that be a violation of the company’s commitment to not punishing failure? ( Kerr & Landauer, 2004 ).
As is so often the case, once these problems surfaced and were declared politically acceptable to talk about, the resolution was fairly straightforward, and turned out to be consistent with Levinson’s recommendations (given that the stretch goals still came from Welch). Norms were developed, and mechanisms established, to require and permit regular interactions among all the parties who would be affected by a particular stretch goal. While admittedly not given equal weight, when future stretch goals were set by Welch, all the affected parties were informed and given an opportunity to voice their concerns and opinions before the goals were implemented.
In this chapter, we have tried to present an objective overview of the concept of stretch goals. Our personal view is that the criticisms of stretch goals are fair and accurate, but in the main, they describe weaknesses in the way that stretch goals (indeed, goals in general) are typically conceived and implemented. Awareness of these weaknesses is a vital precursor to using them but should not be a deterrent.
There’s an old saying that “if you do what you’ve always done, you’ll get what you’ve always gotten.” While the risks of using stretch goals on a widespread basis in an organization are real, so are the risks of not doing so. Most of us do not have readily available alternatives that promote risk taking, catalyze innovation, and stimulate people to come up with startlingly different ways to get things done. For all their risks and flaws, there is abundant evidence that, in a wide variety of circumstances, stretch goals have been successful at achieving such results.
In this regard, Jack Welch’s own conclusion about his company’s experiences with stretch goals bears repeating. He said “we have found that by reaching for what appears to be impossible, we often actually do the impossible; and even when we don’t quite make it, we inevitably wind up doing much better than we would have done” ( Subhnij, 2008 ).
Beyond the specific deliverables due to particular stretch goals, a wider point also seems worth making:
Whatever their magnitude, the specific quantities of the goals that are set invariably turn out to be less important than the conversations, energy, and activities they stimulate The immediate purpose of stretch goals in GE was to try to achieve the designated targets (without punishing failure), but the broader purpose was to get employees to conceive of their jobs and perform their tasks in fundamentally different and innovative ways. The message that GE was trying to convey was that most people typically access only a small portion of their creative energy—that most people have a huge capacity to do things quicker, better, and cheaper.
(Kerr & Landauer, p. 136)
Among the numerous events sponsored by GE-Crotonville in the 1990s were a series of “innovation days.” At one of these events, one of the consultants staged a demonstration in which he gave a team of six or seven people an orange and told them that each person had to touch the orange, but only one at a time. He said that they could throw it to each other or do anything else they wanted, but the orange had to end up in the hands of the person who touched it first.
The team started throwing the orange back and forth, and the consultant timed it. The first time they did it in about nine seconds. When asked to improve, they stood a little closer, threw a little faster, and got it down to about seven seconds.
Then the consultant told them: “Many groups do this in less than a second, and it’s possible to do it in less than half a second.” After a short planning session, the team did it in less than a second, by stacking their hands. The person with the orange dropped it. It went swoosh through everybody’s hands, and the first person caught it at the bottom ( Kerr, 1997 ).
Bartlett, C. A. 1999 . GE compilations: Jack Welch 1981–1999 . Video, Harvard Business School, Boston .
Choo, A. S. Impact of a stretch strategy on knowledge creation in quality improvement projects. IEEE Transactions on Engineering Management, February 2011, Vol. 58, Issue 1, pg. 87 – 96 .
Cyert, R. M., & March, J. G. A behavioral theory of the firm . Englewood Cliffs, NJ: Prentice-Hall, 1963. The Economist. A work-out for corporate America. Academic OneFile. Web. June 28, 2011 .
General Electric . New Manager’s Manual . GE Company, 1984 .
Harry, M., & Schroeder, R. Six Sigma: The breakthrough management strategy revolutionizing the world’s top corporations . Doubleday: NY, 2005 .
Kerr, S. Ultimate Rewards: What Really Motivates People to Achieve ., Boston, MA: Harvard Business School Press, 1997 .
Kerr, S., & Landauer, S. Using stretch goals to promote organizational effectiveness and personal growth: general electric and Goldman Sachs. Academy of Management Executive, 2004 . Vol. 18, No. 4.
Levinson, H. Management by whose objectives? Harvard Business Review, January 2003 .
Litvan, L.M. Leaders & success General Electric’s Jack Welch . Investors Daily . Los Angeles: August 20, 1997 .
Locke, E. A., Latham, G. P. Goal setting: A motivational technique that works . Englewood Cliffs, NJ: Prentice Hall, 1984 .
Oxnard, T. Stretch: How Toyota reaches for big goals . Supply Chain Management Review, March 2004 .
Reinertsen, D. The design factory. Drastic stretch goals can do more harm than good. Electronic Design . June 12, 2000 .
Sherman, S. Stretch goals: The dark side of asking for miracles. Fortune, November 13, 1995 .
Sitkin, S. B., See, K. E., Miller, C. C., Lawless, M. W., & Carton, A. M. The paradox of stretch goals: Organizations in pursuit of the seemingly impossible. Academy of Management Review, 2011, Vol. 36, No. 3, 544 – 566, 22 pgs.
Subhnji, T. Winning the new-generation workforce . Bangkok Post . Bangkok: May 16, 2008 .
Thomerson, L. D. Six Sigma intensified—is 99% good enough? Annual Quality Congress Pro ceedings . Milwaukee: 2002 .
T&D Guru. Monkey business . Training and Development in Australia . Surry Hills: Apr 2011 . Vol. 38, Issue 2; pg. 25, 1pg. www.1000ventures.com/business_guide/goals_stretch.html/
Welch, J. F. Lectures at GE-Crotonville, 1990 – 91, unpublished.
Welch, J. F. Winning . New York: Harper Business, 2005 .
Williams, R. Why “stretch” goals are a waste of time . Financial Post, FP Entrepreneur blog, August 26, 2010 .
Williams, R. Why goal setting doesn’t work . Psychology Today, Wired for Success blog, April 11, 2011.