To Thine Own Self Be True Sustaining superior performance requires knowing what should change and what should stay the same

Case-based analysis suggests that although each of three types of strategic change—Position, Market and Competency—can succeed, changing Position brings with it the greatest risks, while changing Markets and Competencies are typically more successful paths to enduring performance. These observations can help managers see more clearly both the risks and rewards of the different types of strategic change available to them, and to assess their options accordingly.

When and how a company must change in order to sustain superior performance is an evergreen topic of the art and science of management. And like just about every other question of substance, actionable truth is often lost in a vast wasteland of vacuous aphorisms. As but one example, consider that “if it ain’t broke, don’t fix it,” certainly rings true, yet one is simultaneously exhorted to “do it to yourself before someone else does it to you.”

A common synthesis is to argue that when change is required the key is to remain true to some set of core values. Everything should be on the table, we’re told, except that. Sadly, it’s an answer that merely raises another question. How is one to know what those values are—what Polonius in Hamlet called our “self” to which we must be “true”? To speak of preserving core values is to give only the illusion of specificity, for it amounts to saying “don’t change what you shouldn’t.” If we want to make the pursuit of corporate success more predictable, we need something more objective and measurable—and far less circular.

Based on our research into the behaviors of companies with superior performance, we believe it is possible to say something more about the kinds of stability and change that are systematically associated with success or failure. Our findings are not conclusive and our prescriptions are not completely quantitative, but we hope you will find here the seeds of a more transparent, more scientific and less philosophical approach to the pursuit of long-term competitive success.

To begin with our conclusion, we have observed three categories of behavioral change by top-performing companies:

  • Positioning: Changing from differentiated to low cost, or vice versa. This type of change is pursued relatively frequently, but typically with disastrous outcomes.
  • Markets: Changing or expanding markets served by moving into one or more of new products, new geographic regions or new segments.
  • Competencies: Changing or expanding core competencies, most often by reinventing processes that had been critical to the success of a given Position.

These three categories are neither mutually exclusive (companies can change along more than one dimension) nor collectively exhaustive (there are other ways to think about change). Rather, they are empirically derived from our study of the patterns of change exhibited by superior long-term performers.

Changes in Positioning, whether or not they imply changes in Markets or Competencies, typically fail. When they work they take the form of movements “upmarket” to a differentiated position, rather than “down market” to a low-cost position—again, regardless of the level of change in Markets or Competencies required. In contrast, changes in Markets or Competencies typically succeed, regardless of the degree of change required. Changes in Markets tend to be more frequent, however, perhaps because they require less fundamental change: It’s easier to sell what you’re selling now to someone else than it is to reinvent how you do something core to your existing Positioning.

Unearthing these findings has been a two-stage process. First, we needed to identify companies with superior long-term performance (see inset below). To that end, we assess a company’s performance annually, as measured by return on assets (ROA), compared to all publicly traded companies. Companies that finish in the top 10 percent (that is, the 9th decile) often enough that there is a less than 10 percent chance they achieved that result by chance alone are called “Miracle Workers.”

Second, within the Miracle Worker categorization we looked at three trajectories of performance: those that “lost it”—that is, had a strong string of 9s followed by a meaningful string of 8s or less; those that “found it”—the mirror image of “lost it;” and those that appear to have “kept it”—that is, their string of 9s appears to continue to this day.

Identifying superior performers

Ten percent of any population will be in the top 10 percent. That doesn’t mean those “top performers” are any different from the rest of the population, though. To be truly exceptional, a company’s performance has to be better than we would have expected by chance alone.

It turns out that frameworks for competitive success built inductively on a foundation of case study research are often relying on companies with performance profiles indistinguishable from those of lucky random walkers.

The good news is that companies with truly exceptional performance are out there; you just have to know how to look for them.

See to download A Random Search for Excellence for a more complete explanation.

It is by comparing the responses of these three categories of Miracle Workers to competitive and environmental change that we observed that Positioning is the “core value” companies should preserve—the “true self”—while Markets and Competencies are the levers of change associated with long-run success.


There are, at the most fundamental level, two generic positions available in any market: cost leadership and product differentiation.1 Either of these positions can be a foundation for superior profitability. With this categorization scheme in mind, consider now the fates of two Miracle Workers in the “lost it” category: Thomas & Betts Corporation (T&B) and Maytag Corporation (Maytag). Each of these enjoyed sufficient (for the purpose of our analysis) success from the mid-1960s to the mid-1980s thanks to their success as differentiators, yet endured subsequent long runs of less impressive returns as a result of attempting to become cost leaders.

T&B manufactured electrical wiring products of the sort used in residential and commercial construction, while Maytag (which was acquired in 2006) made washing machines and dryers. Strategically the two firms were quite similar. T&B was consistently in the 9th decile for almost 20 years thanks to its innovative products across a wide range of categories—everything from cable ties to electrical junction boxes. Its commitment to innovation showed up in a patent portfolio about double its nearest competitor. Its commitment to customers showed up in a willingness to invest in differentiating seemingly commoditized and low-dollar products. The combined result was a material pricing premium across a wide range of products that showed up as a significant return on sales advantage.

Modest but profitable growth came entirely organically and included measured expansion into international markets where its differentiated strategy translated well: Europe, Canada, Japan and Australia. In contrast, other players in the industry undermined their positions with large acquisitions and premature moves into emerging markets.

Maytag, the appliance manufacturer, similarly owed its early success to a differentiation strategy. The company’s products were perceived to be of superior quality and durability, a perception that was bolstered by a long-term and highly effective advertising campaign built around “Ole Lonely,” the Maytag repairman who never gets a call thanks to the reliability of Maytag appliances. The company translated this brand into a strong pricing premium through significant support for its network of over 10,000 independent retailers. For example, Maytag covered the costs of shipping its products to its distributors, reducing its distributors’ expenses. As a result of this and other such initiatives, distributors were willing to encourage customers to purchase Maytag over the competition. The result was over 20 years of industry-leading performance.

T&B’s decline began coincident with the 1981 recession. There had not been any noticeable shift in the company’s strategy or tactics, so it seems reasonable to conclude that the drop in performance was caused by the general slowdown in economic activity.

In response, T&B embarked on a major change in Positioning. A series of large acquisitions made electronic components almost half of total revenue by 1991, while its historical electrical manufacturing activities were seen as old news. By itself, this would constitute merely a change in Markets, but the electronics business was characterized by standardized technologies that T&B could not differentiate (as it had in the electrical business) and a proliferation of foreign competitors with strong cost advantages. The result was that success demanded strong cost leadership, something T&B proved unable to establish. It wasn’t until a change of senior management in the early 2000s that T&B returned to its roots as a differentiator in electrical components—and saw its ROA begin to recover.

For Maytag, too, the 1981 recession seems to have been a watershed but for different reasons. Cost pressures pushed Maytag and the industry generally into a greater reliance on national retailers. The emergence of “big box” chains in the late 1980s accelerated this trend. Between 1985 and 1996 the number of independent distributors carrying Maytag products fell from over 10,000 to 650.

Relative change? Absolutely!

We have attempted to apply the same rigor to identifying changes in performance as we have exceptional performance. Specifically, by applying a statistical pattern recognition algorithm to each of a firm’s ROA and decile performance we can calculate the likelihood that any given year marks a break with past performance.

This is important, for it allows us to distinguish between a downward trend in ROA—a firm’s “absolute” performance”—and consistency in its decile rank, or its relative performance. Declines in the former might be a signal something is amiss, but maybe not:  Highly profitable positions can erode for all sorts of reasons. Far less ambiguous is how a firm is doing relative to its relevant peer group. When there is reason to believe that this is trending downward, aggressive action is warranted.

Unlike T&B, Maytag remained committed to the appliance industry, but like T&B, attempted to establish a new Position as a broad spectrum, cost-competitive appliance manufacturer. Buying Magic Chef in 1986, at the time a company half its size, moved Maytag into the “mass market” segment of household appliances.2 An even bigger deal followed in 1988 when Maytag almost doubled in size, further diversified its product portfolio and increased its geographic footprint by acquiring Hoover, the UK-based vacuum cleaner company, in 1988.3Unfortunately, Maytag’s performance, both absolute and relative, eroded steadily, to the point that the company was acquired by a much larger and globalized competitor.

In drawing this conclusion, it is important to note that we did not single out T&B and Maytag for study because their repositioning efforts were unsuccessful. Rather, we chose to study them initially simply because of their lifetime performance: They are bona fide Miracle Workers. Further analysis revealed that their Miracle Worker status had a particular profile: a streak of statistically significant high performance, followed by a statistically significant streak of lower performance.

What we have discovered is that high performing companies with this “lost it” profile tend to have responded to adverse events by attempting to change their Positioning. Not every “lost it” company we studied responded this way, and not every repositioning attempt failed. But the grain of the wood is clear: A change in Position is a low-odds proposition.

Figure 1. Performance profile for two "Lost It" Miracle Workers

Markets and Competencies

Miracle Workers that either “found it” or “kept it” often tend to have opted for dramatic change in Markets or Competencies, frequently in order to maintain their historical Positioning, typically as differentiators.

Take, for example, Heartland Express Incorporated, a truckload (TL) trucking transportation services provider. Heartland’s performance since going public in 1985 until 2007 was an essentially unbroken string of 9s: the dip in 1992 was due to a large acquisition, while the 8th decile ranking in 2005 is too isolated to warrant explanation.4 In absolute terms, however, the company has two distinct eras of performance; our statistical analysis identifies 1994 as the change point.

What we have discovered is that high performing companies with this “lost it” profile tend to have responded to adverse events by attempting to change their Positioning. Not every “lost it” company we studied responded this way, and not every repositioning attempt failed. But the grain of the wood is clear: A change in Position is a low-odds proposition.

We have identified three defining elements of Heartland’s strategy. First, the company was historically focused on a relatively small geographic footprint and a small number of customers. Even within those constraints, Heartland remained highly selective with respect to the freight it carried, picking only the most profitable loads.

Second, Heartland was very nearly unique in pursuing this approach because it had the discipline to accept the trade-offs it implied: By focusing in this way, Heartland had to live with much more volatile performance and lower levels of growth than other trucking companies.

Third, Heartland was able to maintain the service levels required to make this strategy work in part thanks to having broken a long-standing industry trade-off. Trucking firms that use owner/operator (o/o) drivers have perforce a smaller asset base than firms that employ their drivers and so must invest in trucks. Traditionally, however, o/o’s have been less reliable than employed drivers. As a result, the lower asset base associated with an o/o fleet typically comes at the price of lower levels of customer service, and hence lower prices. Heartland, however, enjoyed the lower asset base of an o/o fleet without the negative impact on service, and hence pricing, thanks to a richer and more comprehensive compensation package that included not only higher pay but also, among other benefits, full scholarships to university for the children of its drivers. By paying more it was able to maintain a stable and high-performing workforce even though the economic benefits did not accrue entirely to Heartland’s drivers.

Changes in absolute performance, then, can be misleading: Declines might not signal that anything needs fixing, just as increases might not mean you’re doing anything right. Instead, the key to long-term survival seems to lie in knowing when material change is required in order to preserve one’s relative performance position.

During this initial period of higher absolute performance Heartland’s returns were steadily deteriorating. Shouldn’t that have signaled to the company’s leadership that remedial action was required? Based on Heartland’s response, the answer seems to be “yes … and no.” Heartland’s ROA in the 1980s was almost three times the industry average, a level that few efficient markets will long support. One quite reasonably could expect an erosion of profitability as new players entered the market in an attempt to get a piece of the action themselves or as customers vertically integrated in order to reduce their costs.

Changes in absolute performance, then, can be misleading: Declines might not signal that anything needs fixing, just as increases might not mean you’re doing anything right. Instead, the key to long-term survival seems to lie in knowing when material change is required in order to preserve one’s relative performance position.

This is a distinction Heartland seems to have been able to make. During era 1, the company’s advantage from a differentiated strategy built in large part on an o/o fleet was gradually undermined due to increased competition. This shows up in a steadily declining ROA. By the middle of era 1, Heartland began increasing its percentage of employee drivers. However, this was only part of a transformation of the company’s competencies. Rather than resist the increased asset intensity implied by the shift to employee drivers, Heartland also invested heavily in maintaining a new and highly efficient trucking fleet. It went even further, building up and maintaining a much higher trailer-to-truck ratio than other carriers in order to ensure higher levels of service and preserve what it could of its historical
pricing premium.

In order to keep these assets more nearly fully utilized, Heartland invested in growth and in 1994 doubled in size—and expanded its geographic footprint considerably—through the acquisition of Munson Transportation.5 At the same time, it did not let this growth compromise its historical discipline in selecting only the freight that was the most profitable, whatever the short-run impact on asset utilization might be.6 This metamorphosis put an end to the decline in performance and established a new, stable trajectory that defines era 2, which runs from 1994 to today.

It is worth underlining that, absent these changes, Heartland’s streak of exceptional performance would have ended long ago. Extrapolating the slope of performance degradation from era 1 into era 2 finds Heartland in the red by 2002, with an average ROA through 2010 fully 14.4 percentage points per year lower than what it actually achieved.

In short, Heartland had changed its Markets (by shifting to a larger geographic footprint and a more diverse customer base) and some of its Competencies (greater asset intensity and a new human resources model), while preserving others (freight selectivity). In so doing, it was able to maintain the service levels that defined its Positioning, which in turn supported a pricing premium that drove a return on sales advantage. The company’s absolute lead over its competition was noticeably slimmer in era 2 than in era 1, thanks largely to structural shifts in the industry itself. But to the extent that a sustained performance advantage is possible, Heartland seems to have captured it.

Figure 2. Performance profiles for a "Found It" and "Kept It" Miracle Worker

Less dramatic change can deliver equally significant outcomes. Consider Linear Technology Corporation, which became a public company in 1985. A designer and manufacturer of high-performance, highly customized analog microprocessors, its primary customer was the U.S. government, including the Department of Defense (DoD). As the provider of a differentiated product with only weak substitutes, Linear was well positioned to command a price premium. However, since the DoD was the only customer for those products, that pricing power had its limits, and Linear’s profitability, although admirable, was not nearly what it would be within a decade.

The key to Linear’s breakthrough was taking its show on the road and providing similarly highly differentiated products to a variety of new markets and geographic regions. By 2005 government work had fallen from nearly half of total sales to less than 3 percent, and overseas markets went from much less than half to well over 70 percent of total sales. This didn’t happen automatically: Linear had to continue to invest heavily in R&D in order to provide those markets what they needed, and the reduced reliance on defense work exposed the company to a sharp contraction in revenue in 2001. But Linear’s long-term commitment to its historical Position served it well.

What Heartland and Linear reveal in very different ways is the power of persistence, of sticking with a Position that a company understands well and can likely implement more effectively than one it must attempt to master on the fly. Where one might argue that Heartland had evidence that its strategy was successful and worth pursuing, surely the degradation in absolute ROA must have given pause: A far less precipitous fall provoked Maytag into a dramatic repositioning attempt. How was Heartland to know that the key to its success was Positioning and not its Market focus or particular unique Competencies? And Linear’s performance was hardly notable at first: Its upward trajectory is only obvious in hindsight. How was it to know that its Positioning was sound and that it would be able to transplant its Competencies to a broader array of Markets, both product and geographic, with such remarkable success?

All that live must die

The simple answer is that they couldn’t have, any more than T&B and Maytag could have known that their attempts to change Position were going to fail. The data—in the form of performance, competitive analysis, the attractiveness of new opportunities, and so on—are never conclusive. Every company that plies the turbulent waters of strategic change chooses a course that makes sense at the time. In short, it’s not readily apparent ex ante what distinguished successful from failed attempts at strategic change. Unfortunately, it’s rarely readily apparent.

Securing and keeping a competitive advantage is the result of myriad daily decisions, many of which resist meaningful codification beyond general principles. The quest for hard and fast rules is quixotic. The trick is therefore not to be “objective” but to have the “right bias,” and thankfully, the data do offer some guidance. Our case-based analysis suggests strongly that although each of our three types of strategic change—Position, Market, Competency—can succeed, changing Position brings with it the greatest risks, while changing Markets and Competencies are typically more successful paths to enduring performance. These observations, we hope, help managers see more clearly both the risks and rewards of the different types of strategic change available to them and to assess their options accordingly.

We must accept the fact that nothing lasts forever. The great companies we have identified will eventually run aground, even if they persist and change in all the right ways. It is nevertheless more than a vain hope that an understanding of what has led to the demise of once great companies and how others have achieved or sustained their success—so far—can give us some insight into what it takes for the rest of us to achieve a greater longevity and vigor. If we can heed Polonius and be true to ourselves we might at least defer meeting up with Yorick in the boneyard.

Note:  Our model for assessing the streaks of high and low relative performance is probabilistic. At a 90 percent confidence level, Heartland’s streak of 9th decile performances ends in 2008, thanks to four 8s in six years. At a 99 percent confidence interval, however, the streak is still “on,” and these 8s are plausibly a statistically insignificant “blip.”  In contrast, for the “lost it” Miracle Workers, their streaks have ended to a near certainty:  Maytag has ceased to exist as an independent company and so no recovery is possible;T&B could once again deliver a string of 9s, but the string of lower-performing years is sufficiently long that all T&B can do now is begin a new streak, not re-establish a pre-existing one.


Read the book:  The Three Rules: How Exceptional Companies Beat the Odds, by Raynor and Ahmed, by Portfolio, the business imprint of Penguin Group (USA).