human & organizational learning: executive leadership ......neurial start-up. senior executives...

Human & Organizational Learning: Executive Leadership Program Writing Sample

The writing sample activity is designed as part of the application process to provide applicants the opportunity to demonstrate their written communication capabilities, which are critical for doctoral studies. Applicants should prepare a response to the question below in narrative format, developing sentences and/or paragraphs rather than bullet points. To complete the writing sample requirement, applicants are asked to read the attached article (Christensen.Raynor.2003) and respond to the following question:

Why do the authors think it important for practitioners to know what makes a good theory? Do you agree or disagree with their reasons; why?

HARVARD BUSINESS REVIEW

Managers often make significant business decisions based onlittle more than convincing book jacket blurbs. They should holdthemselves-and the experts-to a higher standard.

WhyHard-NosedExecutivesShould CareAbout

I MAGINE GOING TO YOUR DOCTOR because you're notfeeling well. Before you've had a chance to describeyour symptoms, the doctor writes out a prescription

and says,"Take two of these three times a day, and call menext week."

"But -1 haven't told you what's wrong," you say." Howdo I know this will help me?"

"Why wouldn't it?" says the doctor. "It worked for mylast two patients."

No competent doctors would ever practice medicinelike this, nor would any sane patient accept it if they did.Yet professors and consultants routinely prescribe suchgeneric advice, and managers routinely accept such ther-apy, in the naive belief that if a particular course of actionhelped other companies to succeed, it ought to helptheirs, too.

Consider telecommunications equipment provider Lu-cent Technologies. In the late 1990s, the company's threeoperating divisions were reorganized into 11 "hot busi-nesses." The idea was that each business would be runlargely independently, as if it were an internal entrepre-neurial start-up. Senior executives proclaimed that thisapproach would vault the company to the next level ofgrowth and profitability by pushing decision making downthe hierarchy and closer to the marketplace, thereby en-abling faster, better-focused innovation. Their belief wasvery much in fashion; decentralization and autonomy

ManagementTheory

by Clayton M.Christensen and Michael E. Raynor

SEPTEMBER 2003 m

Why Executives Should Care About Management Theory

appeared to have helped other large companies. And thestart-ups that seemed to be doing so well at the time wereall small, autonomous, and close to their markets. Surelywhat was good for them would be good for Lucent.

It turned out that it wasn't. If anything, the reorgani-zation seemed to make Lucent slower and less fiexible inresponding to its customers' needs. Rather than savingcosts, it added a whole new layer of costs.

How could this happen? How could a formula thathelped other companies become leaner, faster, and moreresponsive have caused the opposite at Lucent?

It happened because the management team ofthe dayand those who advised it acted like the patient and thephysician in our opening vignette. Tbe remedy they used-forming small, product-focused, close-to-the-customerbusiness imits to make tbeir company more innovativeand flexible-actually does work, when business units areselling modular, self-contained products. Lucent's leadingcustomers operated massive telephone networks. Theywere buying not plug-and-play products but, rather, com-plicated system solutions whose components had to beknit together in an intricate way to ensure that theyworked correctly and reliably. Such systems are best de-signed, sold, and serviced by employees who are not hin-dered from coordinating their interdependent interac-tions by being separated into unconnected units. Lucent'smanagers used a theory that wasn't appropriate to theircircumstance-with disastrous results.

Theory, you say? Theory often gets a bum rap amongmanagers because it's associated with the word "theoret-ical," which connotes "impractical."But it shouldn't. A the-ory is a statement predicting which actions will lead towhat results and why. Every action that managers take,and every plan they formulate, is based on some theoryin tbe back of their minds that makes them expect theactions they contemplate will lead to the results theyenvision. But just like Monsieur Jourdain in Moli^re'sLe Bourgeois Gentilhomme, who didn't realize he had beenspeaking prose all his life, most managers don't realizethat they are voracious users of theory.

Good theories are valuable in at least two ways. First,they help us make predictions. Gravity, for example, isa theory. As a statement of cause and effect, it allows usto predict that if we step off a cliff we will fall, withoutrequiring that we actually try it to see what happens.Indeed, because reliable data are available solely aboutthe past, using solid theories of causality is the only way

Clayton M. Christensen is the Robert and Jane Cizik Profes-sor of Business Administration at Harvard Business Schoolin Boston. Michael E. Raynor is a director with Deloitte Re-search and a professor at the Richard Ivey School of Busi-ness in London, Ontario, Canada. This article elaborates ona central theme of their forthcoming book, The Innovator'sSolution (Harvard Business School Press, 2003).

managers can look into the future with any degree of con-fidence. Second, sound theories help us interpret thepresent, to understand what is happening and why. The-ories help us sort the signals that portend importantchanges in the future from the noise that has no strategicmeaning.

Establishing the central role that theory plays in man-agerial decision making is the first of three related objec-tives we hope to accomplish in this article. We will also de-scribe how good theories are developed and give an ideaof how a theory can improve over time. And, finally, we'dlike to help managers develop a sense, when they read anarticle or a book, for what theories they can and cannottrust. Our overarching goal is to help managers becomeintelligent consumers of managerial theory so that thebest work coming out of universities and consulting firmsis put to good use-and the less thoughtful, less rigorouswork doesn't do too much harm.

Where Theory Comes FromThe construction of a solid theory proceeds in threestages. It begins with a description of some phenome-non we wish to understand. In physics, the phenomenonmight be the behavior of high-energy particles; in busi-ness, it might be innovations that succeed or fail in themarketplace. In the exhibit at right, this stage is depictedas a broad foundation. That's because unless the phenom-enon is carefully observed and described in its breadthand complexity, good theory cannot be built. Researcherssurely head down the road to bad theory when they im-patiently observe a few successful companies, identifysome practices or characteristics that these companiesseem to have in common, and then conclude that theyhave seen enough to write an article or book about howall companies can succeed. Such articles might suggest thefollowing arguments, for example:

• Because Europe's wireless telephone industry was sosuccessful after it organized around a single GSM stan-dard, the wireless industry in the United States wouldhave seen higher usage rates sooner if it, too, had agreedon a standard before it got going.

• If you adopt this set of best practices for partneringwith best-of-breed suppliers, your company will succeedas these companies did.

Such studies are dangerous exactly because they wouldhave us believe that because a certain medicine hashelped some companies, it will help all companies. To im-prove understanding beyond this stage, researchers needto move to the second step: classifying aspects ofthe phe-nomenon into categories. Medical researchers sori: dia-betes into adult onset versus juvenile onset, for example.And management researchers sort diversification strate-gies into vertical versus horizontal types. This sortingallows researchers to organize complex and confusing

68 HARVARD BUSINESS REVIEW


phenomena in ways that highlight tbeir most meaningfuldifferences. It is then possible to tackle stage three, whichis to formulate a hypothesis of what causes the phenom-enon to happen and why. And that's a theory.

How do researchers improve this preliminary theory,or hypothesis? As the downward loop in the diagrambelow suggests, the process is iterative. Researchers usetheir theory to predict what they will see when they ob-serve further examples of the phenomenon in the vari-ous categories they had defined in the second step. If tbetheory accurately predicts what tbey are observing, theycan use it with increasing confidence to make predictionsin similar circumstances.'

In their further observations, however, researchersoften see something the theory cannot explain or predict.

Prediction Formation of a theory:A statement of whatcauses what and why

i.Categorization

Observation and descriptionofthe phenomenon

Confirmation

Anomaly

an anomaly that su^ests something else is going on. Theymust then cycle back to the categorization stage and addor eliminate categories - or, sometimes, rethink thementirely. The researchers then build an improved theoryupon the new categorization scheme. This new theory stillexplains the previous observations, but it also explainsthose that had seemed anomalous. In other words, thetheory can now predict more accurately how the phe-nomenon should work in a wider range of circumstances.

To see how a theory has improved, let's look at the wayour understanding of intemational trade has evolved. Itwas long thought that countries with cheap, abundantresources would have an advantage competing in indus-tries in which such resources are used as important inputsof production. Nations with inexpensive electric power,for example, would have a comparative advantage in mak-ing products that require energy-intensive productionmethods. Those with cheap labor would excel in labor-intensive products, and so on. This theory prevailed untilMichael Porter saw anomalies the theory could not ac-count for. Japan, with no iron ore and little coal, becamea successful steel producer. Italy became the world's dom-inant producer of ceramic tile, even though its electricitycosts were high and it had to import much of the clay.

Porter's theory of competitive clusters grew out of hisefforts to account for these anomalies. Clusters, he postu-lated, lead to intense competition, which leads compa-nies to optimize R&D, production, training, and logistics

processes. His insights did not mean that prior notionsof advantages based on low-cost resources were wrong,merely that they didn't adequately predict the outcome inevery situation. So, for example, Canada's large pulp andpaper industry can be explained in terms of relativelyplentiful trees, and Bangalore's success in computer pro-gramming can be explained in terms of plentiful, low-cost, educated labor. But the competitive advantage thatcertain industries in Japan, Italy, and similar places haveachieved can be expiained only in terms of industry clus-ters. Porter's refined theory suggests that in one set of cir-cumstances, where some otherwise scarce and valuableresource is relatively abundant, a country can and shouldexploit this advantage and so prosper. In another set ofcircumstances, where such resources are not available,

policy makers can encourage the develop-ment of clusters to build process-based com-petitive advantages. Governments of nationslike Singapore and Ireland have used Porter'stheory to devise cluster-building poiicies thathave led to prosperity in just the way his re-fined theory predicts.

We'll now take a closer look at three aspectsof the theory-building process: the impor-tance of explaining what causes an outcome(instead of just describing attributes empiri-cally associated with that outcome); the pro-

cess of categorization that enables theorists to move fromtentative understanding to reliable predictions; and theimportance of studying failures to building good theory.

Pinpointing CausationIn the early stages of theory building, people typicallyidentify the most visible attributes ofthe phenomenon inquestion that appear to be correlated with a particularoutcome and use those attributes as the basis for catego-rization. This is necessarUy the starting point of theorybuilding, but it is rarely ever more than an important firststep. It takes a while to develop categories that capturea deep understanding of what causes the outcome.

Consider the history of people's attempts to fly. Early re-searchers observed strong correlations between beingable to fiy and having feathers and wings. But when hu-mans attempted to follow the "best practices" of the mostsuccessful fiyers by strapping feathered wings onto theirarms, jumping off cliffs, and flapping hard, they were notsuccessful because, as strong as the correlations were, thewould-be aviators had not understood the fundamentalcausal mechanism of flight. When these researchers cate-gorized the world in terms of the most obvious visibleattributes of the phenomenon (wings versus no wings,feathers versus no feathers, for example), the best theycould do was a statement of correlation - that the posses-sion of those attributes is associated with the ability to fiy.

SEPTEMBER 2003

why Executives Should Care About Management Theory

Researchers at this stage can at best express their find-ings in terms of degrees of uncertainty: "Because such alarge percentage of those with wings and feathers can fiywhen they fiap (although ostriches, emus, chickens, andkiwis cannot), in all probability I will be able to fiy if I fab-ricate wings with feathers glued on them, strap themto my arms, and fiap hard as I jump off this cliffT'Thosewho use research still in this stage as a guide to actionoften get into trouble because they confuse the correla-tion between attributes and outcomes with the under-lying causal mechanism. Hence, they do what they thinkis necessary to succeed, but they fail.

A stunning number of articles and books about man-agement similarly confuse the correlation of attributesand outcomes with causality. Ask yourself, for example, ifyou've ever seen sttidies that:

• contrast the success of companies funded by venturecapital with those funded by corporate capita! (implying

that the source of capital iimding is a cause of successrather than merely an attribute that can be associatedwith a company that happens to be successful for somecurrently unknown reason).

• contend that companies run by CEOs who are plain,ordinary people earn returns to shareholders that are su-perior to those of companies run by flashy CEOs (imply-ing that certain CEO personality attributes cause com-pany performance to improve).

- assert that companies that have diversified beyondthose SIC codes that define their core businesses returnless to their shareholders than firms that kept close to

their core (thus leaping to the conclusion that the attri-butes of diversification or centralization cause share-holder value creation).

• conclude that 78% of female home owners betweenthe ages of 25 and 35 prefer this product over that one(thus implying that the attributes of home ownership,age, and gender somehow cause people to prefer a spe-cific product).

None of these studies articulates a theory of causation.All of them express a correlation between attributes andoutcomes, and that's generally the best you can do whenyou don't understand what causes a given outcome. In thefirst case, for example, studies have shown that 20% ofstart-ups funded by venture capitalists succeed, another50% end up among the walking wounded, and the rest failaltogether. Other studies have shown that the success rateof start-ups funded by corporate capital is much, muchlower. But from such studies you can't conclude that your

start-up will succeed if it is funded by venture cap-ital. You must first know what it is about venturecapital-the mechanism-that contributes to a start-up's success.

In management research, unfortunately, manyacademics and consultants intentionally remain atthis correlation-based stage of theory building inthe mistaken belief that they can increase the pre-dictive power of their "theories" by crunching hugedatabases on powerful computers, producing re-gression analyses that measure the correlations ofattributes and outcomes with ever higher degreesof statistical significance. Managers who attemptto be guided by sucb research can only hope thatthey'll be lucky - that if they acquire the recom-mended attributes (which on average are associatedwith success), somehow they too will find them-selves similarly blessed with success.

The breakthroughs that lead from categoriza-tion to an understanding of fundamental causali^generally come not from crunching ever more databut from highly detailed field research, when re-searchers crawl inside companies to observe care-fully the causal processes at work. Consider theprogress of our understanding of Toyota's produc-

tion methods. Initially, observers noticed that the stridesJapanese companies were making in manufacturing out-paced those of tbeir counterparts in the United States.The first categorization efforts were directed vaguely to-ward the most obvious attribute -that perhaps there wassomething in Japanese culture that made the difference.

When early researchers visited Toyota plants in Japanto see its production methods (often called "lean manu-facturing"), though, they observed more significant attri-butes of tbe system-inventories that were kept to a min-imum, a plant-scheduling system driven by kanban cardsinstead of computers, and so on. But unfortunately, they



leaped quickly from attributes to conclusions, writingbooks assuring managers that if they, too, built manufac-turing systems with these attributes, they would achieveimprovements in cost, quality, and speed comparable tothose Toyota enjoys. Many manufacturers tried to maketheir plants conform to these lean attributes - and whilemany reaped some improvements, none came close toreplicating what Toyota had done.

The research of Steven Spear and Kent Bowen has ad-vanced theory in this field from such correlations by sug-gesting fundamental causes of Toyota'sability to continuaily improve quaiity,speed, and cost. Spear went to work onseveral Toyota assembly lines for sometime. He began to see a pattern in theway people thought when they designedany process-those for training workers,for instance, or installing car seats, ormaintaining equipment. Erom this care-ful and extensive observation. Spear andBowen concluded that all processes at Toyota are de-signed according to four specific rules that create auto-matic feedback loops, which repeatedly test the effec-tiveness of each new activity, pointing the way towardcontinual improvements. (Eor a detailed account of Spearand Bowen's theory, see"Decoding the DNA ofthe ToyotaProduction System," HBR September-October 1999.)Using this mechanism, organizations as diverse as hospi-tals, aluminum smelters, and semiconductor fabricatorshave begun achieving improvements on a scale similar toToyota's, even though their processes often share few vis-ible attributes with Toyota's system.

Moving Toward PredictabilityManned flight began to be possible when Daniel Ber-noulli's study of fiuid mechanics helped him understandthe mechanism that creates lift Even then, though, un-derstanding the mechanism itself wasn't enough to makemanned fiight perfectly predictable. Eurther research wasneeded to identify the circumstances under whicb tbatmechanism did and did not work.

When aviators used Bernoulli's understanding to buildaircraft with airfoil wings, some of them still crashed.They then had to figure out what it was about those cir-cumstances that led to failure. They, in essence, stoppedasking the question, "What attributes are associated withsuccess?" and focused on the question, "Under what cir-cumstances will the use of this theory lead to failure?"Theylearned, for exampie, that if they climbed too steeply,insufficient lift was created. Also, in certain types of tur-bulence, pockets of relatively lower-density air formingunder a wing could cause a sudden down spin. As aviatorscame to recognize those circumstances tbat required dif-ferent technologies and piloting techniques and others

Unfortunately, manymanagement researchersare so focused on howcompanies succeed thatthey don't study failure.

that made attempting flight too dangerous, manned flightbecame not just possible but predictable.

In management research, similar breakthroughs in pre-dictability occur when researchers not only identify thecausal mechanism that ties actions to results but go on todescribe the circumstances in which that mechanismdoes and does not result in success. This enables them todiscover whether and how managers should adjust theway they manage their organizations in these differentcircumstances. Good theories, in other words, are circum-

stance contingent: They define notjust what causes what and why, butalso how the causal mechanism willproduce different outcomes in dif-ferent situations.

For example, two pairs of re-searchers have independently beenstudying why it is so difficult forcompanies to deliver superior re-turns to shareholders over a sus-

tained period. They have recently published carefully re-searched books on the question that reach opposingconclusions. Profit from the Core observes that the firmswhose performance is best and lasts longest are, on aver-age, those that have sought growth in areas close to theskills they'd honed in their core businesses. It recom-mends that other managers follow suit. Creative Destruc-tion, in contrast, concludes that because most attractivebusinesses ultimately lose their luster, managers need tobringthe dynamic workings of entrepreneurial capitalisminside their companies and be willing to create new corebusinesses.

Because they've juxtaposed their work in such a help-ful way, we can see that what the researchers actuallyhave done is define the critical question that will leadto the predictability stage of the theory-building cycle:"Under what circumstances will staying close to the corehelp me sustain superior returns, and when will it becritical to set the forces of creative destruction to work?"When the researchers have defined the set of differentsituations in which managers might find themselves rela-tive to this question and then articulated a circumstance-contingent theory, individuals can begin following theirrecommendations with greater confidence that they willbe on the right path for their situation.

Circumstance-contingent theories enable managers tounderstand what it is about their present situation thathas enabled their strategies and tactics to succeed. Andthey help managers recognize when important circum-stances in their competitive envirormient are shifting sothey can begin "piloting their plane" differently to sustaintheir success in the new circumstance. Theories that haveadvanced to this stage can help make success not only pos-sible and predictable but sustainable. The work of build-ing ever-better theory is never finished. As valuable as

SEPTEMBER 2003 71


Porter's theory of clusters has proven, for example, thereis a great opportunity for a researcher now to step in andfind out when and why clusters that seem robust can dis-integrate. That will lead to an even more robust theory ofintemational competitive advantage.

The Importance of FailuresNote how critical it is for researchers, once they have hy-pothesized a causal mechanism, to identify circumstancesin which companies did exactly what was prescribed butfailed. Unfortunately, many management researchers areso focused on how companies succeed that they don'tstudy failure. The obsession with studying successfril com-panies and their "best practices" is a major reason whyplatitudes and fads in management come and go with suchalarming regularity and why much early-stage manage-ment thinking doesn't evolve to the next stage. Managerstry advice out because it sounds good and then discard itwhen they encounter circumstances in which the recom-mended actions do not yield the predicted results. Theirconclusion most often is, "It doesn't work."

The question, "When doesn't it work?" is a magical keythat enables statements of causality to be expressed incircumstance-contingent ways. For reasons we don't fullyunderstand, many management researchers and writersare afraid to tum that key. As a consequence, many apromising stream of research has fallen into disuse anddisrepute because its proponents carelessly claimed itwould work in every instance instead of seeking to leamwhen it would work, when it wouldn't, and why.

In a good doctor-patient relationship, doctors usuallycan analyze and diagnose what is wrongwith a specific patient and prescribe anappropriate therapy. By contrast, the re-lationship between managers, on theone hand, and those who research andwrite about management, on the other,is a distant one. If it is going to be use-fril, research must be conducted andwritten in ways that make it possiblefor readers to diagnose their situationthemselves. When managers ask ques-tions like, "Does this apply to my industry?" or "Does itapply to service businesses as well as product businesses?"they really are probing to understand the circumstancesunder which a theory does and does not work. Most ofthem have been burned by misapplied theory before. Toknow unambiguously what circumstance they are in,managers need also to know what circumstances they arenot in. That is why getting the circumstance-defined cate-gories right is so important in the process of building use-ful theory.

In our studies, we have observed that industry-basedor product-versus-service-based categorization schemes

For most managers,trying out a new idea tosee if it works is simpiynot an option: There istoo much at stake.

almost never constitute a useful foimdation for reliabletheory because the circumstances that make a theory failor succeed rarely coincide with industry boundaries. TheInnovator's Dilemma, for example, described how pre-cisely the same mechanism that enabled upstart compa-nies to upend the leading, established fimis in disk drivesand computers also toppled the leading companies in me-chanical excavators, steel, retailing, motorcycles, and ac-counting software. The circumstances that matter to thistheory have nothing to do with what industry a companyis in. They have to do with whether an innovation is or isnot financially attractive to a company's business model.The mechanism-the resource allocation process-causesthe established leaders to win the competitive fightswhen an innovation is financially attractive to their busi-ness model. And the same mechanism disables themwhen they are attacked by disruptive innovators whoseproducts, profit models, and customers are not atfractiveto their model.

We can trust a theory only when, as in this example, itsstatement describing the actions that must lead to suc-cess expiains how they will vary as a company's circum-stances change. This is a major reason why the world ofinnovating managers has seemed quite random-becauseshoddy categorization by researchers has led to one-size-fits-all recommendations that have led to poor results inmany circumstances. Not until we begin developing the-ories that managers can use in a circumstance-contingentway will we bring predictable success to the world ofmanagement.

Let's return to the Lucent example. The company isnow in recovery: Market share in key product groups has

stabilized, customers report increasedsatisfaction, and the stock price is recov-ering. Much ofthe tumaround seems tohave been the result, in a tragic irony,not just of undoing the reorganizationof the 1990s but of moving to a stillmore centralized structure. The currentmanagement team explicitly recognizedthe damage the earlier decentralizationinitiatives created and, guided by a the-ory that is appropriate to the complexity

of Lucent's products and markets, has been working hardto put back in place an efficient stmcture that is alignedwith the needs of Lucent's underlying technologies andproducts.

The moral of this story is that in business, as in medi-cine, no single prescription cures all ills. Lucent's manag-ers felt pressured to grow in the 1990s. Lucent had a rela-tively centralized decision-making structure and its fairshare of bureaucracy. Because most of the fast-growingtechnology companies ofthe day were comparatively un-encumbered with such structures, management con-cluded that it should mimic them - a belief not only en-



dorsed but promulgated by a numbeT of managementresearchers. What got overlooked, with disastrous conse-quences, was that Lucent was emulating the attributes ofsmall, fast-growing companies when its circumstanceswere fundamentally different The management neededa theory to guide it to the organizational structure thatwas optimal for the circumstances the company was ac-tually in.

Becoming a Discerning Consumerof TheoryManagers with a problem to solve wil! want to cut to thechase: Which theory will help them? How can they tell agood theory from a bad one? That is, when is a theory suf-ficiently well developed that its categorization scheme isindeed based not on coincidences but on causal links be-tween circumstances, action, and results? Here are someideas to help you judge how appropriate any theory or setof recommendations will be for your company's situation.

• When researchers are just beginning to study aproblem or business issue, articles that simply de-scribe the phenomenon can become an extremelyvaluable foundation for subsequent researchers'attempts to define categories and then to explainwhat causes the phenomenon to occur. For exam-ple, early work by Ananth Raman and his col-leagues shook the world of supply chain studiessimply by showing that companies with even themost sophisticated bar code-scanning systems hadnotoriously inaccurate inventory records. Theseobservations led them to the next stage, in whichthey classified the types of errors the scanning sys-tems produced and the sorts of stores in whichthose kinds of errors most often occurred. Ramanand his colleagues then began carefully observingstocking processes to see exactly what kinds of be-haviors could cause these errors. From this foun-dation, then, a theory explaining what systemswork under what circumstances can emerge.

- Beware of work urging that revolutionarychange of everything is needed. This is the fallacyof jumping directly from description to theory. Ifthe authors imply that their findings apply to allcompanies in all situations, don't trust them. Usu-ally things are the way they are for pretty goodreasons. We need to know not only where, when, and whythings must change but also what should stay the same.Most ofthe time, new categorization schemes don't com-pletely overturn established thinking. Rather, they bringnew insight into how to think and act in circumstance-contingent ways. Porter's work on intemational compet-itiveness, for example, did not overthrow preexisting tradetheory but ratber identified a circumstance in which a dif-ferent mechanism of action led to competitive advantage.

• If the authors classiiy the phenomenon they're de-scribing into categories based upon its attributes, simplyaccept that the study represents only a preliminary steptoward a reliable theory. The most you can know at thisstage is that there is some relationship between the char-acteristics of the companies being studied and the out-comes they experience. These can be described in terms ofa general tendency of a population (20% of all companiesfunded by venture capital become successful; fewer ofthose funded by corporate capital do). But, if used toguide the actions of your individual company, they caneasily send you on a wing-fiapping expedition.

• Correlations that masquerade as causation often takethe form of adjectives -/7um6/e CEOs create shareholdervalue, for instance, or venture-capital funding helps start-ups succeed. But a real theory should include a mecha-nism-a description of how something works. So a theoryof how funding helps start-ups succeed might suggest thatwhat venture capitalists do that makes the difference ismeter out small amounts of funds to help the companies

feel their way, step by step, toward a viable strategy. Fund-ing in this way encourages start-ups to abandon unsuc-cessful initiatives right away and try new approaches.What corporate capitalists often do that's less effective isto flood a new business with a lot of money initially, al-lowing it to pursue the wrong strategy far longer. Thenthey pull the plug, thus preventing it from trying differentapproaches to find out what will work. During tbe dot-com boom, when venture capitalists fiooded start-ups

SEPTEMBER 2003 73


with money, the fact that it was venture money per sedidn't help avert the predictable disaster.

• Remember that a researcher's findings can almostnever be considered the final word. The discovery of a cir-cumstance in which a theory did not accurately predict anoutcome is a triumph, not a failure. Progress comes fromrefining theories to explain situations in which they pre-viously failed, so without continuing our examination offailure, management theory cannot advance.

When Caveat Emptor Is Not EnoughIn shopping for ideas, there is no Better Business Bureaumanagers can turn to for an assessment of how usefula given theory will be to them. Editors of managementjournals publish a range of different views on importantissues-leaving it to the readers to decide which theoriesthey should use to guide their actions.

But in the marketplace of ideas, caveat emptor-lettingthe reader beware - shirks the duty of research. For mostmanagers, trying out a new idea to see if it works is sim-ply not an option: There is too much at stake. Our hope isthat an understanding of what constitutes good theorywill help researchers do a better job of discovering the

mechanisms that cause the outcomes managers careabout, and that researchers will not be satisfied with mea-suring the statistical significance of correlations betweenattributes and outcomes. We hope they will see the valuein asking, "When doesn't this work?" Researching thatquestion will help them decipher the set of circumstancesin which managers might find themselves and then framecontingent statements of cause and effect that take thosecircumstances into account.

We hope that a deeper understanding of what makestheory useful will enable editors to choose which piecesof research they will publish - and managers to choosewhich articles they will read and believe - on the basis ofsomething other than authors' credentials or past suc-cesses. We hope that managers will exploit the fact thatgood theories can be judged on a more objective basis tomake their "purchases" far more confidently. V

1. Karl Popper asserted that when a researcher reaches the phase in which atheory accuiateiy predicts what has been observed, the researcher can stateonly that the test or experiment"feiied to disconfirm" the theory. See The Logicof Scientific Discovery (Harper & Row, 1968).

Reprint R0309D; HBR OnPoint 4937To order, see page 135.

"We already have quite a few people who know how to divide,so essentially, we're now looking for people who know how to conquer."


Harvard Business Review Notice of Use Restrictions, May 2009

Harvard Business Review and Harvard Business Publishing Newsletter content on EBSCOhost is licensed for

the private individual use of authorized EBSCOhost users. It is not intended for use as assigned course material

in academic institutions nor as corporate learning or training materials in businesses. Academic licensees may

not use this content in electronic reserves, electronic course packs, persistent linking from syllabi or by any

other means of incorporating the content into course resources. Business licensees may not host this content on

learning management systems or use persistent linking or other means to incorporate the content into learning

management systems. Harvard Business Publishing will be pleased to grant permission to make this content

available through such means. For rates and permission, contact [email protected].

human & organizational learning: executive leadership ......neurial start-up. senior executives...

Documents