If, like me, you enjoyed Jim Collins and Jerry Porras’s business bestseller Built to Last, if you instinctively embraced its message that successful companies are built on strong cultures, core values, and a mission that goes beyond making money, then you might want to avoid The Halo Effect, Phil Rosenzweig’s important new book. Although if you do, you’ll be missing out on an intriguing, clear-eyed business book and one that will help public relations people understand the stories that are told about their companies and clients.

Built to Last is not the only one of my favorite books skewered in The Halo Effect. Rosenzweig also demolishes the evidence presented by John Kotter and James Heskett in Corporate Culture and Performance, which I have been quoting approvingly for more than a decade. And a host of other works, from In Search of Excellence to Good to Great get the same treatment. And the author throws in convincing critiques of the Fortune Most Admired Corporations list and its Best Companies to Work For research for good measure.

Rosenzweig doesn’t actually disprove any of the claims made by Built to Last or In Search of Excellence or any of the other works he examines, but he does demonstrate that the rigorous research methodologies described by their authors are fatally flawed, that their conclusions—although they may well be valid—are not supported by the evidence presented, and that almost everything we thought we knew about the causes of business success, we don’t.

The problem with all of these books, and with the vast majority of business journalism, is The Halo Effect, a flaw in the observational process that makes it extremely difficult, if not impossible, to distinguish cause and effect when analyzing corporate success. Rosenzweig’s argues convincingly that the companies held up as role models by Collins and Kotter and Peters and Waterman didn’t succeed because they had great cultures; rather, they had great cultures—or cultures perceived as great—because they were successful.

“This central idea of this book is that our thinking about business is shaped by a number of delusions,” Rosenzweig writes. Those delusions “distort our understanding of company performance [and make] it difficult to know why one company succeeds and another fails. These errors of thinking pervade much that we read about business, whether in leading magazines or scholarly journals or management bestsellers. They cloud our ability to think clearly and critically about the nature of success in business.”

The starting point for his case is a fascinating social science experiment conducted by a professor from the University of Illinois, in which groups of participants were asked to estimate a company’s future sales and earnings per share based on a set of financial data. After the experiment, some groups were told they had performed well, while others were told they had performed badly—but he did so at random, so that some of the actual high performers were told they had done badly while some of the poorer performers were told they had done well.

Then he asked the participants to assess how well their groups had done on a range of issues. Those who were told they had done well—even if they had, in fact, performed badly—described their groups as highly cohesive, with better communication, more openness to change, and superior motivation. Those who were told they had done badly—even if they had, in fact, performed well—recalled a lack of cohesion, poor communication, low motivation.

In other words, people attribute positive characteristics to groups that are effective, and negative characteristics to groups that are ineffective—but they do so after the fact, when the results are known. It’s not hard to see how this “halo effect” could apply in the corporate world, leading academics, authors and journalists—and even managers within the organizations themselves—to attribute positive cultural characteristics to companies that outperform the market and positive leadership qualities to the CEOs who lead them.

On the corporate culture front, Rosenzweig points to the example of Cisco, which went public in 1990 and 10 years later reached a market capitalization of $555 billion, briefly overtaking Microsoft to become the most valuable company in the world.

Business magazines began to ask what had made Cisco such a standout success. They came up with a few themes that were repeated in almost all of the coverage of the company: the leadership of CEO John Chambers, who was humble and customer focused; Cisco’s remarkable skill of acquiring companies (it was “an acquisition engine… cleverly designed and highly tuned” to the Internet market, wrote Fortune) and ability to integrate those acquisitions (with a “very organized, methodical approach… [to] managing the experience of acquired employees); it understood how to motivate employees (a Wired article focused on the company’s “shiny, happy people).

Then the dot-com bubble burst. In May of 2000, Fortune asked on its cover whether John Chambers was the world’s best CEO and inside argued that if you owned stock in just one company, Cisco was the one to own. By the end of the year, those shares, which had been worth $80 each, had declined to around $38. In early 2001, the company wrote off $2.2 billion in inventory, having overestimated demand. And so, exactly one year after its original cover story, Fortune wrote another. Not surprisingly, the tone was very different.

Where once the magazine’s writes had seen a relentless focus on the customer, now they saw “a cavalier attitude” and sales techniques that were “irksome,” alienating competitors. Business Week followed up, claiming that what had previously been seen as a disciplined acquisition strategy and skilful integration of those acquisitions was a “buying binge” characterized by “haphazard” and “freewheeling” investment practices.

Says Rosenzweig: “Read in the context of their times, each of these articles seems plausible. They offer a reasonable explanation of events. But look at them over the course of a few years and we have to question whether the reporters got the story right—or if their descriptions were colored by the stories they wanted to tell.

“Facts were assembled and shaped to tell the story of the moment, whether it was about great performance or collapsing performance or about rebirth and recovery. Placing these accounts together, the impression is nothing short of Orwellian—a rewriting of history that thrusts facts into the past, rearranging the record to tell a more coherent story. It’s an example of reinterpreting the past to suit present needs.”

That’s no just a story about journalistic hyperbole, it’s also a symptom of a deeper problem we all have in understanding corporate performance.

“For all the attention that Cisco received,” Rosenzweig says, “for all its prominence in the press over several years, even experienced journalists and respected academics had trouble identifying with any precision the reasons for Cisco’s outstanding success or its stunning decline. There was talk, over and over, about customer orientation and leadership and organizational efficiency, but these things are hard to measure objectively, so we tend to make attributions about them based on thing we do feel certain  about—revenues and profits and share price.”

If that’s a tendency in evaluating corporate success, it’s an even greater factor when it comes to judging CEO performance, as Rosenzweig demonstrates when he looks at the career of ABB chief executive Percy Barnevik, who was named Europe’s most admired CEO and became the subject of numerous articles and business books. When the Financial Times named ABB Europe’s Most Respected Company for the third year running in 1996, it hailed Barnevik’s “strategic vision and focus,” while John Micklethwait and Adrian Wooldridge of The Economist wrote in their book The Witch Doctors that the ABB chief was “over-endowed with energy” and one of Europe’s “management superstars.”

But after Barnevik stepped aside and ABB’s fortunes began to decline, the tone of the coverage changed. Says Rosenzweig: “When ABB was posting record performance, Barnevik had been the focus of a virtual personality cult, portrayed with powers of a superman. He had been described as charismatic, bold, and visionary. But once performance fell, Barnevik was remembered as arrogant, imperial, and resistant to criticism.”

In this shift, Rosenzweig detects another common fallacy, the tendency to attribute company success to the skills or shortcomings of a single individual.

“One of the main reasons we love stories is that they don’t simply report disconnected facts but make connections about cause and effect, often ascribing credit or blame to individuals. Our most compelling stories often place people at the center of events. When times are good, we lavish praise and create heroes. When things go bad, we lay blame and create villains.”

Thus, the leaders of successful companies are always said to have a handful of important—but hard to define or quantify—characteristics: clear vision, effective communication skills, self-confidence, personal charisma. “You can always find good things to say about leaders at successful companies,” says Rosenzweig, “and you can always find reasons to criticize leaders of failing firms.”

As for the business books, some cynics will already have questioned the predictive value of In Search of Excellence and Built to Last based on the post-publication performance of many of the profiled companies. The 35 “excellent” companies profiled by Tom Peters and Robert Waterman that were publicly trade, for example, underperformed the S&P 500 over the subsequent five and 10 years. The companies featured in Built to Last included HP, Motorola, Merck and Disney, all of which encountered significant difficulties in the coming years. More than half of those “visionary” companies failed to match the S&P 500 over the next five years.

“Lasting business success,” Rosenzweig concludes, “is largely a delusion… We can tell ourselves that a handful of companies, selected by rigorous and objective process, are a breed apart, somehow better than the rest… But we’re kidding ourselves. If we start with the full data set and look objectively at many years of corporate performance, we find the dominant pattern is not one of enduring performance at all, but one of rise and fall, of growth and decline.”

Success is not random, he says, but “chance does play a role, and the difference between a brilliant visionary and a foolish gambler is usually inferred after the fact, an attribution based on outcomes.”

And success is not exclusively dependent upon the merits of a company’s actions, because those actions must always be viewed in relation to the actions of competitors. “Companies can never achieve success simply by following a given set of steps, no matter how well intentioned; their success will always be affected by what rivals do….

“That’s an uncomfortable truth, because it admits that some elements of business performance are outside our control. It’s far more appealing to downplay the relative nature of performance or ignore it completely. Telling a company it can achieve high performance, regardless of what competitors do, makes for a more attractive story.”

Similarly, the Fortune Most Admired list shows signs of a halo effect. The scores on almost all criteria are driven to a significant extent by a company’s financial performance, so that when Cisco’s financial results declined, so too did its scores for innovation, for people, for social responsibility. The company also saw its rating on the Best Companies to Work For index decline.

“It’s logical to think that having satisfied employees ought to lead to high performance,” says Rosenzweig. “After all, satisfied employees might be willing to work harder and longer, and might care more about keeping their customers happy. It sounds right.”

But such conclusions might be confusing cause and effect. “The challenge is to untangle the direction of causality. Does lower employee turnover lead to higher company performance? Perhaps, since a company with a stable workforce might be able to provide more dependable customer service, spend less on training, and so forth. Or does higher company performance lead to lower employee turnover? That could be true as well, since a profitable and growing company might offer a more stimulating and rewarding environment as well as greater opportunities for advancement.”

The point is that numerous business books and articles have simply assumed the former, when in reality “knowing which leads to which is critical if managers want to know what to do—how much they should invest in greater levels of satisfaction versus other objectives.”

If so many business articles and books are built around a fallacy (or, more appropriately, since Rosenzweig identifies a number of fallacies, several), why do they continue to sell so well when business people are supposedly so hard-nosed about hard measures of success?

“Managers are busy people, under enormous pressure to deliver higher revenues, greater profits, and ever larger returns for shareholders,” says Rosenzweig. “They naturally search for ready-made answers, for tidy plug-and-play solutions that might give them a leg up on their rivals. And the people who write business books—consultants and business school professors and strategy gurus—are happy to oblige. Demand stimulates supply, and supply finds a ready demand.”

Rosenzweig cites the physicist Richard Feynman, who defined science as “a method for trying to answer questions which can be put in the form: ‘If I do this, what will happen.’”

Plenty of business questions lend themselves to scientific experimentation, Rosenzweig says. “Imagine you want to know where to place an item in a supermarket, or what effect a price change will have on the quantity of a product sold, or what effect a special promotion will have…. In fact, just about any situation with an abundance of similar transactions affords a natural setting for experiments….

“But other questions in business don’t easily lend themselves to this sort of experimentation. Take a major strategic initiative, like the launch of a new product. Coca-Cola didn’t get two chances to launch New Coke in 1985. It got one bite at the apple and famously got it completely wrong… There’s simply no way to bring the rigor of experimentation to questions like these.”

He examines the strategic decision by Danish toy manufacturer Lego to expand beyond building blocks into other areas, including licensed Harry Potter figures. When the company’s revenues and profits declined in 2004, it fired its chief operating officer and business reports were united in their explanation: the company “had strayed too far from its core. It lost sight of its roots.” Similar criticisms were leveled at British retailer WH Smith, which had “strayed” or “drifted” from its original focus on newspapers and magazines to become a more broad-based retailer.

Yet when Nokia successfully expanded beyond mobile phone handsets to expand into mobile gaming, imaging, music, and even complex wireless systems for corporations—all appealing because of their promise of growth and high margins—the company was applauded. “So why didn’t Business Week say that Nokia was straying or drifting? Why was Nokia merely expanding? Because at the time, no one knew if Nokia would succeed or fail, so Business Week chose a nice neutral verb, expand…. If Nokia could make this change work, it would be celebrated for its nimble strategy and clever management. Of course, if Nokia failed, reporters would say it had erred by moving into areas it didn’t understand; it had strayed or drifted.”

In other words, we tend to create stories to explain the success or failure of strategic decisions, but we do so retroactively, after we know the outcomes of those decisions. So those stories are of almost no value in helping others make better decisions in the future.

Here, Rosenzweig distinguishes between stories and reports. “A report is above all responsible for providing the facts, without manipulation or interpretation. If the accounts about Lego and WH Smith are meant to be reports—which presumably they are, since they’re written by reporters—then words like stray and drift are problematic.

“Stories, on the other hand, are a way that people try to make sense of their lives and their experiences in the world. The test of a good story isn’t its responsibility to the facts as much as its ability to provide a satisfying explanation of events.”

In Search of Excellence, Rosenzweig says, “worked wonderfully as a story. It told the tale of successful American companies prevailing against stiff competition. It focused managers of a few key points: people, customers, action. It provided inspiration.”

Rosenzweig is not necessarily hostile to the conclusions reached by In Search of Excellence of Built to Last or any of the other books he critiques. “I’m sympathetic to much of what these bestsellers have to say,” he writes. “Most companies can benefit from their basic principles—and if they authors have been able to present them in a way that’s accessible and vivid, so that millions of managers read these ideas and take them to heart, perhaps that’s not such a bad thing.”

But the notion that a company can choose to be great—and its corollary, that a troubled company is only troubled because of its failure to implement the ideas contained in these books—is both delusional and dangerous. “We’re not far off from self-help books that tell people they can be millionaires in five easy steps or lose 20 pounds in two weeks, or awaken the power of greatness in them.”

These books “may divert our attention from a more powerful insight—that while we can do many things to improve our chances of success, at its core business performance retains a large measure of uncertainty. Business performance may actually be simpler than it is often made out to be, but may also be less certain and less amenable to engineering with predictable outcomes.”

At the very least, The Halo Effect will help professional communicators understand the stories that are told—in the media and in business literature—about their companies, and to understand that many of those stories are based on fallacious thinking. But it can also help public relations people understand the kind of stories they should be telling, and the kind of expectations they should—or should not—be raising.

Finally, it’s worth noting that Rosenzweig is himself a wonderful storyteller, despite his skepticism about the power of storytelling to bend the facts to fit the moral. The Halo Effect is a quick, easy and compelling read.