ࡱ> q` dbjbjqPqP .t::\%$$$8\p$ $j""Ln(.$0$0$0$0$0$0$$%h(vT$"""T$i$###"X.$#".$### &$"X#.$$0$#y(2#dy(##hy($ (h#TT$T$#$"""" D D f^ On the origin of strategies Evolution across a population is nature's trick for mastering uncertainty. Businesses can use it too. ERIC D. BEINHOCKER In 1988, as I wandered around the floor of Comdex, the computer industrys vast annual trade show, I could feel the anxiety among the participants. Since the birth of the IBM PC, six years earlier, Microsofts Disk Operating System (DOS) had been the de facto standard for PCs. But DOS was now starting to age. Everyone wanted to know what would replace it. Apple Computer, at the peak of its powers, had one of the largest booths, showcasing the brilliantly graphical Macintosh operating system, which made DOS look antique. Meanwhile, two different alliances of major companies, including AT&T, Hewlett-Packard, and Sun Microsystems, offered graphical versions of the UNIX operating system. And IBM, determined not to let Microsoft overtake it again, was touting its new OS/2, an operating system said to combine DOS compatibility with the power of UNIX and the Macs ease of use. Amid the uncertainty, there was something very curious about the Microsoft booth, which was hardly the shows largest. Despite the success of the company, almost all of its competitors were much bigger. But while most booths focused on a single blockbuster technology, Microsofts resembled a Middle Eastern bazaar. In one corner, the company was previewing the second version of its much delayed and highly criticized Windows system, which had as yet gained little market share. In another, Microsoft touted its latest release of DOS. Elsewhere, it was displaying OS/2, which it had developed with IBM. In addition, Microsoft was demonstrating major new releases of Word, Excel, and other applications that ran on Apples Mac; indeed, they were the most popular programs for it. Finally, in a distant corner, Microsoft displayed SCO UNIX, a PC-compatible version of the UNIX operating system, developed by a firm that had a marketing agreement with Microsoft. "What am I supposed to make of all of this?" grumbled a corporate buyer standing next to me. Columnists wrote that Microsoft was adrift, that its chairman and chief operating officer, Bill Gates, had no strategy. Reporters told stories of infighting at the company as one group of programmers worked furiously on Windows and DOS while others poured their energies into OS/2, Mac applications, and SCO UNIX. In the face of uncertain markets, Microsoft followed the only robust strategy: betting on every horse Although the outcome of this story is now well-known, to anyone standing on the Comdex floor in 1988, it wasnt obvious which operating system would win. In the face of this uncertainty, Microsoft followed the only robust strategy: betting on every horse. Bill Gates wanted Windows to win. Yet if DOS had continued to dominate, he would have continued to supply it. If OS/2 had triumphed, he would have shared the wealth with IBM. Had the Mac emerged victorious, he would have lost the war for operating systems but won in applications. Had UNIX become the standard, Microsoft would have also no longer been the major player in operating systems, but with its hand in SCO UNIX, it could at least have remained a contender. Besides betting on all these horses, Gates took steps that would pay off no matter which of them came in firstinvesting heavily, for example, in graphical-interface design skills and object-oriented programming. The same pattern is evident today in Gatess approach to the Internet. Microsofts ever-shifting portfolio of development projects, investments, and acquisitionsas well as the companys many joint ventures with software, cable, telecommunications, and communications media companieslooks confusing if you ask, "What is Microsofts strategy?" It makes more sense to ask, "What are Microsofts strategies?" Predicting the unpredictable Strategy has traditionally been a high-stakes game that started when a management team developed a vision of the companys future environment. The team then made big, hard-to-reverse decisions about where the company would focus its energies, its capital, and its people. Finally, the team hoped and prayed that its vision was correct. But the whole idea of predicting the future is troublesome. Managers intuitively know that competitive environments are often difficult to foresee. New work by scientists and economists shows that business markets may be even more fundamentally unpredictable than intuition suggests and that they may be subject to the same laws governing other so-called "complex systems," such as galaxies, ecosystems, insect colonies, brains, and the Internet. Recently, scientists have gained a better understanding of these systems, which comprise many dynamically interacting parts. HYPERLINK "https://www.mckinseyquarterly.com/On_the_origin_of_strategies_1060" \l "foot1" 1 An important property of such systems is the inherent difficulty of predicting how they will behave. As our limited success in predicting the course of the weather and the stock market shows, past patterns of behavior are not always reliable guides to the future. The properties of complex adaptive systems present particular challenges to the development of business strategy because people have a natural tendency to look for patterns. Indeed, the human drive to find patterns is so strong that they are often read into perfectly random data. Moreover, human beings like to assume that cause directly precedes effect, which makes it difficult to anticipate the second-, third-, and fourth-order effects of path dependence. If businesses are indeed complex systems, one cannot rely on the ability to find patterns and make predictions. A better way to manage uncertainty is needed. Strategy as evolutionary search For clues about how to develop strategies in an uncertain world, look to another complex system, where strategy is a matter of life and death: the world of living species in their evolutionary fight for survival. Nature teems with examples of ingenious strategies that have proved quite adaptive in complex environments. Evolutionary biologists have developed tools to understand the way nature develops its strategies, and these tools can also help businesspeople think about business strategies. The key to natures ability to develop strategies in the face of uncertainty is the fact that evolution occurs not in individuals but across populations. A species is always a population of strategies, with new adaptations, or experiments, constantly emerging. Likewise, businesses should cultivate evolving populations of strategies, much as Microsoft did in 1988. To understand how populations evolve, scientists use "fitness landscapes": imaginary three-dimensional grids with surfaces of peaks and valleys. HYPERLINK "https://www.mckinseyquarterly.com/On_the_origin_of_strategies_1060" \l "foot2" 2 Each point on the grid represents a possible gene combination of a particular species, and any points height indicates that combinations fitness for survival. Adjacent points represent gene combinations that are only slightly different (Exhibit 1). To understand how to frame strategies in the business world, replace "species" with "company" and "gene combinations" with "business strategies." One point on the grid, for instance, might represent a strategy of "focusing on US customers with a narrow product offering." Another might represent "selling globally with a broad one-stop-shop product line." The height of each point on the grid would represent that strategys profitability, positive or negative.  INCLUDEPICTURE "https://www.mckinseyquarterly.com/image/article/chart/chart_orst00_01.gif" \* MERGEFORMATINET  In the evolutionary struggle, species search for the high points on their fitness landscapes, which can assume various forms. Stuart Kauffman, of the Santa Fe Institute and the Bios Group, points to a "Mount Fuji" landscape, with a single steep peak representing a strategy superior to all others. On such a landscape, nearby points (representing similar strategies) have similar altitudes, or levels of fitness. At the other extreme would be a perfectly random landscape of peaks and valleys, where there is no one optimal strategy, and neighboring points, representing similar strategies, might have radically different levels of fitness. Most complex systemsbiological and commercial alikedont embody either extreme: they may include many peaks and valleys, like random landscapes, but nearby points tend to have similar levels of fitness, like the Mount Fuji example. Such a "rugged" landscape has not only hills and valleys but also regions of alpine highlands and coastal lowlands. Of course, both biological and business landscapes are not static; they buckle and heave as competitive conditions shift. Surviving in the fog To carry the analogy further, a business strategist can be regarded as the leader of an expedition to find the highest elevations on a companys fitness landscape. Fog prevents the "hikers" from seeing more than a few feet in front of themselves, and the landscape keeps changing. Biologists have identified key rules that evolution uses to find high peaks. Managers too can use them. The first rule is that evolutionary searches never sit still. No matter how successful a strategy is at a given moment, a species or a business must experiment constantly to find something better. In nature, genes are constantly subject to mutation and recombination. In business, the expedition should always be on the move to find new directions upward. Another key principle is parallelism: the entire expedition should not explore the same region. Instead, many search parties should spread out from the "base camp" to explore the shifting terrain and bring back news of discoveries. In nature, evolutionary searches are massively parallel, for each member of the population is a unique experiment searching through a slightly different point on the landscape. The more places the search parties explore, the more likely they are to find a new, higher peak (Exhibit 2), so the more explorers go out searching, the more likely it is that someone will have located a good spot when the site of the expeditions base camp starts collapsing. Finally, the deployment of many small search parties lets the expedition pursue a few high-risk hunches because each party is somewhat expendable.  INCLUDEPICTURE "https://www.mckinseyquarterly.com/image/article/chart/chart_orst00_02.gif" \* MERGEFORMATINET  A good example of business parallelism in action comes from the highly successful credit card company Capital One, which routinely generates large numbers of new ideas, tries them out in the marketplace, backs the winners, and unsentimentally kills off the losers. HYPERLINK "https://www.mckinseyquarterly.com/On_the_origin_of_strategies_1060" \l "foot3" 3 The company, for instance, is using its skills in data mining and direct marketing to resell mobile phonesa risky experiment that will quickly help it determine if this is an attractive new spot in the landscape. But most companies resist parallelism, which can be expensive, seems inefficient, and sometimes puts people at cross-purposes by creating internal competition. Of course, randomly sending out dozens of search parties is unwise, for no matter how large the expedition, there is likely to be too much ground. Some of the explorers should therefore embark on what mathematicians and biologists call an "adaptive walk," or "hill climbing," by taking a step, peering through the fog for the steepest uphill path, taking a step in that direction, looking around again, taking another step, and so on. In biological evolution, genetic mutation provides that sort of incremental variation. This approach works well in regions where high ground tends to lie near other high areas. But once adaptive walkers arrive at the topmost point in a vicinity, they will be stuck there, since the next step leads downhill even if a much higher peak looms in the fog just over a nearby valley. Some search parties should therefore follow a different strategy. Suppose they had powerful "pogo sticks" permitting them to hop to more distant points. In biological evolution, sexual reproduction supplies this ability to take longer leaps across a landscape. Unfortunately, though the pogo stick gives the explorers a chance to find higher peaks, the fog makes it hard for them to predict where they will land. Since the highest peaks tend to be near one another, the farther the jump, the greater the chance of losing rather than gaining altitude. Adaptive walks ensure that most of the time the explorers are heading toward a higher level of fitness The best strategy for searching a rugged fitness landscape is to intersperse adaptive walks with occasional medium- or long-distance pogo jumpsan idea that not only can be proved mathematically and in computer simulations but also makes intuitive sense. Adaptive walks ensure that most of the time the explorers head toward a higher level of fitness. Jumps on the pogo stick keep them from getting stuck on local peaks and occasionally yield significant improvements, at the cost of occasional declines in fitness. In nature, a speciess experiments tend to be incremental; most of the population explores ground relatively close to the base camp. Although long-distance jumps involve a smaller part of the population, they provide very valuable protection against sudden shifts in the landscape, for in biological systems, a few genetically atypical individuals may survive a sudden environmental change that kills off typical individuals. The same is true in business. A company should use most of its resources to build its current business, but the resources devoted to riskier experiments further afield are also critically important, since such experiments could contain the seeds of success in a currently unimaginable future. HYPERLINK "https://www.mckinseyquarterly.com/On_the_origin_of_strategies_1060" \l "foot4" 4 The Finnish telecom company Sonera, for example, mixes short and long jumps by making adaptive-walk improvements in its core fixed-line and mobile-telephone businesses while also placing long-jump bets at the frontiers of a market that doesnt yet exist: the intersection of mobile telephony and the Internet. Yes, your company can do this The implicit metaphor and mind-set of strategy has been military: you set a clear objective and then focus all your energies on "taking that hill." In this view, strategy is about setting objectives, narrowing options, and making choices. But complexity theory teaches that, in reality, the hill might not be there tomorrow. In an uncertain world, strategy is really about creating options and opening up new choices, not shutting them down. True, resources are finite, so trade-offs and commitments are inevitable. To continue the mountaineering analogy, decisions have to be made about where to put the base camp each night and where to send search parties. But these decisions are much more likely to be successful in the rugged, foggy landscape if search parties bring back lots of choices and information. Many companies have to make big bets because they havent sent out sufficient numbers of search parties. Companies that avoid this pitfall do things differently; as executives at Capital One put it, "We dont like to make big bets. If we are making a big bet, it probably means we failed somewhere earlier." This emphasis on exploration and opening up options is very different from what most companies actually do. A common objection to the evolutionary approach is the idea that a company cant afford to bet on everything. Maybe Microsoft can make multiple bets, but other companies cant. Actually, in 1988 Microsoft was smaller than just about all of its competitors. Besides, if you were going to make a bet today on what kind of company will eventually win on the Interneta large incumbent technology company or a leading venture capital firmwhich would you bet on? Probably the latter, for though it would have much less money to spend than the incumbent, it would be exploring much more of the terrain. Resources are rarely the constraint on parallelism and experimentation; the real problems are mind-sets and cultures. Instead of asking if you can afford to do something, ask if you can afford not to do it. You might also ask if there isnt a risk that companies will get involved with businesses they dont understand, as many did in the 1970s. That decades diversification craze was about assembling conglomerates of unrelated businesses to diversify industry sector risk. I am not advocating that here. Searching the fitness landscape implies rather that a company should cultivate a portfolio of strategic experiments in a business it knows. Although some of them may be risky and different from the current business, they leverage core skills and assets. As Microsoft, for instance builds its Internet business, it has become involved with media, content, and communications ventures that expand the bounds of its business but also leverage its brand name and its core skills in software while helping it build new ones. Another objection to the evolutionary approach is that it may be fine for businesses like software or credit cards but not for those requiring heavy capital expenditures. How many experiments can an electric utility, for example, afford to run? Actually, in view of the global trend to deregulation, an electric utility cant afford not to conduct many experiments as the landscape buckles and heaves around it. Although it is true that in some businesses the minimum size of experiments may be relatively large, not all search parties, even in such businesses, must be large; there simply has to be a sufficient number of experiments with a sufficient spread across the landscape. However, once a businessparticularly a capital-intensive businessfinds promising vantage points, it must be able to ramp up the resources it devotes to them quickly. Getting an organization to think about strategy in this way isnt easy. It forces people to deal with ambiguity and creates internal tensions as the company simultaneously pursues what may be contradictory paths. Most corporate processes, metrics, and incentives reflect the "take that hill" view of strategy. They must change. Invest in diversity Pursuing a diverse portfolio of strategies requires a diverse mixture of people to contribute ideas. Diversity of experienceas well as of age, sex, race, national origin, and so forthis therefore crucial. General Electric, for instance, deliberately hires employees from a variety of talent pools and then exposes those people to more than one GE business rather than leave them to climb the career ladder in a particular unit. HYPERLINK "https://www.mckinseyquarterly.com/On_the_origin_of_strategies_1060" \l "foot5" 5 Actively cultivating diversity in this way may seemindeed, may beinefficient in the short run, but the dangers of inbreeding are as great in business as in nature. Value strategies as if they were options The metrics most commonly used to evaluate new investments are net present value (NPV), payback period, operating profit, and return on capital. They share a common flaw: failing to account for the option value of each investmentthe possibility that it could create further opportunities in the future. Since there is real value in having a wide array of choices, which provide flexibility in an uncertain world, companies should account for this value. Fortunately, significant progress has been made in giving a companys strategic options values that can be incorporated into business decision making. HYPERLINK "https://www.mckinseyquarterly.com/On_the_origin_of_strategies_1060" \l "foot6" 6 Categorize the mix of strategies Managers should ask whether their strategy portfolios embody a healthy mix of small and long jumps. Strategies should thus be categorized along three dimensions: time (payoffs in the near, medium, or long term), risk (low, medium, or high), and aim (expanding or defending a current business, building a new one that has already been identified, or laying the foundations for potential new businesses). Stress-test your strategies To ensure that a population of strategies covers sufficiently broad and promising regions of the fitness landscape, ask questions that test for robustness. What are the likely scenarios in the industry? Which of them is best for the company, and how can its strategies push the industry in that direction? How would the company adapt to the other most likely scenarios? Which of them, likely or not, presents major threats or opportunities? Is the company preparing itself for highly unlikely scenarios, or does it accept the risk of being unprepared for them? Classical scenario analysis, HYPERLINK "https://www.mckinseyquarterly.com/On_the_origin_of_strategies_1060" \l "foot7" 7 systems dynamics modeling, HYPERLINK "https://www.mckinseyquarterly.com/On_the_origin_of_strategies_1060" \l "foot8" 8 and new frameworks for categorizing and managing uncertainty HYPERLINK "https://www.mckinseyquarterly.com/On_the_origin_of_strategies_1060" \l "foot9" 9 can help a company stress-test its strategy portfolio. Bring the market inside As much as possible, selection pressures on a companys population of strategies should reflect marketplace pressures. In reality, however, investment dollars often flow to projects favored by people who hold political power or who generated yesterdays revenues rather than tomorrows possibilities. To overcome this pattern, Lucent Technologys new internal-venture arm gains external market valuation for some of its internal start-ups by inviting venture capitalists to participate, and it may eventually float pieces of new businesses in initial public offerings. Moreover, many companies protect egos, turf, and careers by clinging to nonperforming strategies. Capital One tries to avoid this problem by shutting down unsuccessful experiments, but it is careful to distinguish between ideas that failed though they were implemented well, on the one hand, and failed people, on the other. Use venture capital performance metrics Many companies use the same performance metrics for a mature factory that makes widgets in Ohio and a start-up Internet operation in India. It makes sense to measure short-jump investments through traditional financial returns such as profits and return on investment, but long-jump investments, like those of venture capitalists, should be measured through market acceptance, meeting milestones, growth, market capitalization, and so forth. History is littered with great companies that failed to move as the landscapes around them changed. They hunkered down in the base camps of their traditional strategies as their formerly high peaks collapsed, and since they had not adequately explored the landscapes around them, they didnt know where to go. By fashioning an evolving population of strategies, managers can improve the odds that they will avoid the strategic Death Valleys and enjoy the rewards found on the strategic peaks. About the Author Eric Beinhocker is a principal in McKinseys London office. This article is adapted from "Robust adaptive strategies" by Eric D. Beinhocker, Sloan Management Review, spring 1999, pp. 95-106. Reprinted by permission of the publisher. Copyright 1999 Sloan Management Review Association. All rights reserved. This version of the article was first published in The McKinsey Quarterly, 1999 Number 4.     PAGE  PAGE 9 %  """"""%%-&0&**0051617181A2B22222>>????:@X@P*PQQ3R4R5R6RRScUdUUUUUUUwWWYYZ?ZZZ[ZֹjzBhhUh 5CJaJhh0JH*jhhUhhCJaJ hhhh0JhI bq%u "%1&(**$a$gd$a$gd$a$gd$a$gd$a$gd$a$gd$a$gddd** ,-0914!6k99;?:@X@A#EHH|KNP*PRSUUwWW$a$gd$a$gd$a$gd$a$gd[ZZZZZZZR[S[T[U[[[#_K_bcccckddddddddddddddddddddddddºººº°°h 50JmHnHu h0Jjh0JUjhUhhh6]hh0JhhCJaJhh0JH*jhhU hh+W[[]#_K_abcdddddddddddddddh]hgd &`#$gd$a$gd$a$gd$a$gd$a$gddddd$a$gd,1h/ =!"#$% zBDdI?   S fA (chart_orst00_01chart_orst00_01.gifbAN-86$exAD0npAN-86$ePNG  IHDR'8PLTEZVZjfjrnrb^d223--.⌌jjkپڰ tu{EFKMNSUV[|}샅ꮲ뎑־͠\^cdfklns}ھީ֦މޙ#Y "%˾պuz]bhmrxejo|}~ۺIt7c\мFJMVZ]zNRT̨ǁ fjj>??yzz⣤օVZVmnm  ƶ``^;;:776QN0'& \[UCB=KJESRM{zu >:,rjQ}tY^XDÕ|se3/$m{_ƨܧ͛ȷ֤g_KUO@E@4ˮqk\42- MG;sqmkif憂 ZVTJFFRNNzvv#""ĺtttddd[[[TTTLLLDDD((( @Q3bKGDH cmPPJCmp0712Hs=IDATx^} \[畧.6IDSo2R: N0z{xa0ESc4dV̺IHZ{4<:(fAvkg6xw06AOtϽN╦St99s' \ dMd&kdbj8 OCƧb'ICA ҙI[H rUz0AjN@!M]ҟ@bb !w!? "R)Fnz #O)S8e &? OypZm;', B^LNWzR퍛HmK482p.#ȬZbܨTp"\tŰIțZd'dy:Z 5JhX:b@`@)7WH:Nq( Xlr'>kl>&8ZUbC YSl6 ]3OU^~l?s =Ǹ3ssOڻ?^KȳޞcO_}7 }k0BgGW x^gU/ ~| G΂:p-{0@Yn~x{)ݟܣ{w,Iu G̃_+>.x._յ/rO;e/|7@ ,{l?<.+S喕=De <}, ȕܲl8PV{}0/0Υ ?wC/|?/ /h<Nߐr?H _(O[ =`g"ۢ AO* ("v 2 ˄πJ? g2 EoQ qs"SVz?E$rHk cTAwN%HyyyoigiMygOM#wIuS4udMD՚^43H./)>yG}' x9J5%K3G]T#ی3-4x{2;w^]r=WԻpNBuqDFdyj|Gd82eKWN57 =-މ&rMh=[Y?\@RC 3TM$a0wP_o"C7 2mDiƴ>q8xme| MdncHmţm'@oJHtyONQPH 6Mqߤ\h}n1p-- acDi7+7şMt+hA)arA҂ꚤ`o0K+Ǒ)ߔ@賺Њt^yJӑ3myXWwR};ymtߕ۟{W'ǯaII3'dHH̵׏L!zdddj*lגƯun|ʗ0$%!XrH֑:vꍤ#!ݵ뻦ES77l8u,y䫈*9 6,<>~bG a+<&}57>(5ep, ܼݬ{*Ow͛Gfjzc ZܫQy:Xdztf FHaxք޸ᅽc C;Hmŵ_N!Չ+sT껵l{LcҼ47J7z+k)hWrкݳ4(*LaƐĈ^]\ѕ4UbZmyMAzWhuդ!jhD / OFI)M =0!MT$`oJ(hcűk*oӊm>2>102SkW X2aOhl,iHhj|ɑ٣[[zQ.ޟ՞yC*$TMJ&owpRwsgKK+KQ's!m*WgԎ_MYZH}[Qz+6]NUF Fߩ L ^MLЦzW )TZӣRx4~_8l+^s|7R=BZm5 6?@۽4PJ-LMf A?˝ٚ rV7MJ|Rsx&ʯ☗[mM/; ӡcm={CԪG 5UW:j=aJhSˁY_-U_QǏ댾M AOզ_sKlJg\bLU;f ? 8r~*)H3Wu5Suޯ@E<]1ꑫc?xGo 7S QDP^ȣخ]E ;~cd2ʼnsſ;~3yXuhrfsW_p쵪GN"?>LUGnRaWq~LN?FJ_؅$|o7أ51D- |O _IK lBj}]]ש]ȩ+鑪?ćOfTT'fd P5`c,ɰ+AW>pzG.MZ|sR6U0eTo9.v Ʃj*F ~ Fv1PL8%-{5C^}[w? c^B&Ƅӛ9ϫ8Nƙ;{(%11]]D+ q_g;ݦ{Qc\sq}ޯ oNHu6"_4gqqq'=z}c.(0l֨E#6 }|_spwz]wת%,̫K3YB2hU#pJJ_6p&e d#M85+Gdsa KFŒuC7@U mj(Pnu^(aH~VTYO%F}R`D/"e0ڔ}f@-SgI+"6[ī-sԕiX>Tx6%PYLjS-6_MS)l?[a껇G}=v}oonA .x$OT(`(?0o)/EJiP}NXRG8r[UCN 1:$Rv;v[+S}zB_fR+2quce &ena{L7[5VBʢy lnkg@jw{N;=m ]'\~OcD&Kd|(T +J6&d1C(`(]/l5]8 &2[gpQޮwf{NTrAC2 aupTciF)c*/ϦNJEzGXbaȕSwy7i]}{[8v{X'քyBBTR7tƨ>`a+Qb ]le}ee%p%mYOZs9(Ig"H*9 #tAfڑ8ɱ+Lik<~,6wv0w:]îѮS2G*^}'}Gokkv?V2g9;"E:] GZy*f, 76tvu Ūk 8;ZA@g;\]hO30;AL?uh6'/gG|+ U8],\PӚNSᰲIknw Z\ųEugVjޥ벽a(wo4@Gwd3*-Xȭrey`B&JV(O3NHQqO[oMY]>Gwhuo!-yk.Zc5w:O\.'JF0iIX. +y uEtfeIŅ$_bhl8=.wu[`c\m ^O'x622Rfq1`슿޾bKWp4j BH+?\)%vn&ĀLfY?[OH뻇q' 4 Z;j>GJmGJTPfЃb RDiLn)gh?dsUǎL ]z>ͽK3噉Pݝ^?el d`*-uxAaT)d˖y_N=N(ߘŭ0bߍn3gv 8{zX%p6@oda *K;-'"U\ˉ>sh'Xľ TJ% U#V(gNSDc0p(a=#>@X~)K}!Nx8'CJ}o?~ ۚmdy^jTg.=j{D+da<<4ۅ >)7YCk4H({CmLrJތ=Lؙn1qǁe<d^nfWX,–0N`I> 21'ۜɛm]Ρ}33l9S|b(lLZ>:#*!]_]ag0< j.fJ͑ɖPR\zNIIB AЫ~F3sj";Vנ>2^pX=m{n>wgSZ!*-bB&:_<m!%Te`5Fj΁IP%ܩS՛|½dlZgJ4y f|wެ.C5DCshݸP8%}xU".s+H3r)MFE/nt3i&j{`Aj~QR!5"{bW.Pۢ $R(A៣xȁBΓU$Ktl`H*5MI$CAZXޥ{$!IUX}pJUQ0lD[L%Vpœt]铊o)hm]4Q9blֳѸZЧ% `f4 }J>#[>(SQ:۩eMcQB,@+da: L8-%e>cF!SݿlsYAmzQg U-|NA=BQNv]qo f^A)I Caa3tW@^NA.|Vbaskn;O@ٔʀa jsgκxL.$B6G;n]r;s~+% q>*c 4ƔC.3ap7uSScfwULI]9b*by"DV^* 7BFÅߋgG;i8ZCk3S˛JE34&J]gjn;kfsx7).Ij5yi> ˈ%܁{S~ ?/, N)Fcjyhtu4{8Ayo{/Vj_M`ǯUQ!@2U Γ5Ņbv&H|`7g=ō띤5Ȗ~mvK_G*dg33$\'M~~N,(&N'cdF 至qf)PL,pC~0٘h|ζ8Oh}Nx}gɫع:CLmkkt XNb% 6$Kt&.U#M El] ݤjkv[*q>7`| YԹԮ'휵|nJ;lbg[{ڃ'Ϳog/M;Q8!#%b|39-Hx8'73L·4^*GO@'cڰ͊2ony5o4&(5Ae(gvP81m63u&SP-Ln(2xeIxz{C)߻{WpF 8!p_)lE%)TxܻSWE%́,iJ1[HgKq'ގSuNJ],5 p{Ŵ5껇zɺNy:B3?T tA+R 72Kr++NrGt:D )-[}Kc )ܔ3EQj*(MO2|Aj@`K@.yVy֠M%d^- 0(Jo)ontz"gi7)ykC 3zzwT0Y 9@em&6@g?PaÇ 8K'A(MM!_b 6%)m N0 $bkb]OD;_D\} \Pi&L%)~Z|Aj%r8l$j/fz=C^ Gmg#YA "ir?.sv{G8cXz_u?caAlՏGDЩEb'RH(WʳQMؓR)+b$p)廡x>Yw#'yɌo>GO_"q'nj<9>"cgd{?Kf_{,WmrP(iTY!T ӃQ[WJLǎԀ2e~/]SeWGgTbi4K [ i/l&TKrD aʲemk`+M;5A ފq6"?n U~ DW}Tn콍@Dш1,n[ dUȕ{/ %}_ L(5z5 7 UNQ-4d*32(.W-f :1D2M T,uymCqMřGznoT'|b͸جHdLR~/Flw} ErHN')7>- '@ʻׅ ;I xnCuJp""h*BaSybJ n &ݷ\o+X"ExYoSyxjyw/0v7D]7{tEya{#²>1>EmesT N9,faM<_(*l&C٪NtI!C6[2/$n%,F}xNZ" zI"4o} Xd8"\u㻄r֑T)=#5h?O3 $DfH+/]p5Z?.]< b!6ggC 1X4?1Q=sl^X.>N`apm}0b'"/ښ,Vh@v8:^7,&Tݟ[ã~ :ZIoD:vv ̭v=pZNXA&(2͇4 ӆF&*>&b9G;yP kD; */4,[$8[fk+0.Y+{_ڳٚ=w gau@!7f {Yh' bxx>(=miquPK|[u h3'WE6)l!H&- e++sD+f_5z 9zLKD%(wYZ2pVEFbno'@xx hNr;mo>Ӽ pp ?f:?HU/HTK0%tgY,N>0~+EQ BsU~b[%x/K)Nx1 2N XZQh&%|(hp5I,۠Ooiw|PAEdB{TO:*\OFj߰/ ">ggL;E o?cg.#M^L}?)m½B@:Cic l$Q҃}z-HP\%c82DJ?ie77D\pA93 wYaY J[aCe}{0R":m)X>O`)[Jܸ=|%_TS(_ÁM'F/Zl+G)vhy!рLKp+)G;D||FDiLV졬&R`pmDcG4 z^ݲ">_%hq/:n8AۆVK0ٲD> X+.\^AF~}zH]2໻;`< Vjkvtw2B jE1 - rNʦ`,UXqýxso, Snj=rΝqPK A>i*0,LyhѲ~ /&@q!_+6 %(;~倰ls9j,o\ OttY=U͉ZNG萨? .*&V(:{Innni?{Imj,ԬMP Ē)&P 8!6P͖<ed2IuSp#uXo}6b]ܙfȇd<ւBɲY"]0Pau"B0=u^g/I6".nUf1HmSs߀3}ma6E(.n(*\a&`|]nOc!)TFhNFT,p_)m?T /nz9qBWJw{l 3F, _1mVF5y8mF`O1R 9HniB?/8ͣjQ^7 P魔hDʬa Rdft: m(ϑ3Q0w Ihx0Q$`:|go~ x)}WSHUSsDrA4z7]@#VIyhR:1xLV8e_3!udqem[#[[3n Ĵaa(z!'Ѥ-ƒs6@1(TuҮL_; ^^pߙ@WFJh!BD}cGg$w5vZ;M{@4UbkDhz! f 8K['|=$tq `E6bZ(޻ s8ci?%KÃbE;qE0(HE<=̃9P5XiuSkk(V˶0BLRob=)Lĵ0d2~gN]UEBf8XW- NT`E[ěZj>#!||_o贑VIᩏT]蝑)1w'B R>y×e BQiPVb Sſ󽶴V D>/S($d6Բ zC{D`#\gj N'%jED4K؋RZM4f[LёxM0_k@ +1޺{+}IJQ#E/9C )Ms pC$[1Sf{iwQF"ߜ需!"R[8v'cD=&ƗeBU&M_˸8%k/Is TG| 9~Sk1JfnU<-g2YR^IV:HD^q4ڥH#*qiT27Z~?6xEԼh16OݑL9u^]68DS 4ѭg*S#J y ^_8H$NԱ vvYZ8l (Vk5k \ou T$c2(/ui50_c\TJ4gUPtdAAo^W\Ԗ{|% !...w@]T0Ec[0ˋD& HN,y`B`8= iڠQDtoD ]'{p7 D*%7+׎VA'\*rOIzf&ʺ_L.i~w rz0"JF,٬їC2^Dž1v$E#~:[%K7V*vDYeJɽ8MF8iureVPCYd7M{.t& aXq rW' =k0fshq򊂴Ӧ2,ۻjCS8_(s,N!Xv]h[ E2^,)2ܻ{Yhx 9 ou=Ňꡡ/^,#5@q4 )H^FP|+o >y/}8@Yf$JL6a1$cX,3N/KդcW1i0/Ȏ!Or%&iǡ$^*THJw \,K_Мe J8Y% %PYP| jPRRM@NF*`;߼e.y2T@ZSdT{3VjULCF4)݈,*CATw-/NG2"dTMGJw@¡3Gfunl()AؕPGD1}p&FN[Fm!-=~_Du#Xe@LpR4Ҽ=/y.54~S.3{N.PFG+_pE"OޯoOqd}YË|Ǔ$tB\Gq7Ty%ʄ٬n<j@k v6)FoߞKSK}+TԄwIOEk𽏓m*J7.,`+N~54?M@JPi|RE ztul݈U'8fOFPtw~ Rn4n-tmO_޽M.aiJ C*>$&\yr-HQ2iF흍!]%Źi ERR2Bϗ3^Q/p}[R:o7u7tՄ}Uw8U5א>.]{X-?E/yZK.j򟴾jⴴ5ijWq UVq f*N!۬UyZ)6b*O8ņ@lVim֪<͢M^]B`r2YGMM:fIENDB`b?Dd  S fA (chart_orst00_02chart_orst00_02.gifb> ^Npw#eTs`>B0nX> ^Npw#eTsPNG  IHDR"^>PLTE۽ڧ̙̜̙ƷЯ͙ஈŴ̙̙̙f̝sf{fff|~f3z]fff3lntffPQfffpfN33ff33f`WC3ffSTY33mBE3PI8Ruf33ABDf3@:-33333f('%f3f3HtRNS`bKGDHgIFg>U cmPPJCmp0712Hs;IDATx^} cֲ-~[Bb y@DjZzQZ3cl8wѽ'iifjmUy[^8???_de#Z<8Ya]7ó?c@ xL>9Qϻh&z/8(zBgxHx7]7H[/yGĨgYDy6-DPh*v e;/C*E?Tp얽\.ju+h.'V*ϡINk?TLuT܁w;a-t=C X-D_2 p p*-[l"E`-4^ٗowux/>g̳+λt=@!_ګWwD?&H_}ۗϸ}۫'`jϭ_>l"&-Lg4Wg^Ͽ*|~^s/ǧ` }3`կy!9K`JS x5@ B;">=_>#J.\D'߁{'r#@7Wѧ/wp#|O}~y>A"Hч߫za%̗CS –@De?H H>#~I>*E[;3Jhe1Eyai\9H}" ik}CNc/@<}"|8{  ¢v,^zՇO0Sggx8pv S (wCWgpF9|O"D~{E2<;MhC|(Dq_/_)eGVR'$Vu-kj耋"FRtPWݽboA;2fOvA+=pK+a4:HvX؎.q(nNUM^d^7+Ƨxp?U 90MfݬCs*DhooboNVTpZ͞p$?b]=0n`DmC(OU6O]VsQ3T hYYDic=q 3mzLl0*%QSe n%p]Ӭ,Hr" D$7\@#z.\qDi$w roO 4%Gr4^]mECV) Џ_@d>u TmGbq b/@^|'nȕr2Ϋrx}tM:HzY̢ a͘lRz0.yp8)Q&=/<!~K^^@}6u+ѭjD@9H1G,Zs sxøD$C@{1F񺄧$T/5#E/$_!}Ndv$0N0kxuյwM"@5^z+4 Ƭ-.u-`hۅٟe屢x8C1@rj^ "2Ǩ\YImimiƏ"EeWID ^~ $r{K 5]"۹񲅤a%_!k HG٪JKRO7c29_z9NRbc.C?8˅E1krf&F}T'P Y"IVӬ%&'?vа _ro$P wg sw~u6HEau${isڳ' %r&ʓ s;q[tJDQ |)uvORb38X~v-S.ɡ ̍ʰ}IdLvAA>Kd8kj`{O*`E8TRQou;Q;bD`H^9)I2&|v8}->]%?j s!Xi)uEp_bx׸=>G" ѼIeb4lYv~Т?FH p @u.`؝ǔRd$|MlŐP$r +_'SBmE0RT i,?A0K`sk&۫aG*>r!DM~"@0dG\?&9DqS.'-z_ OH(ΛzQ/gxDM".,_i¡uSjp;n zU5`B#h`W42ĤfHj+3u{1Xu\yxW1U I,& `Lz8PݍA/qQbsU^T<*կQMGy!AME%MYI  3*?1k-R$# Q&S;E9~fItD$*AdY3vp8] -}-^5̠iPSG-vk=&(Ĥ4&AVDEO@2<ÑPDA C תJR ZyD w lqɩfm,H0IKj~I^o͆#iϊ P >|\`1,A'5.umd%OOAyyck|b@Sr. Gd_MJh, y#~UiE\z_2+`"5,!#:Sͫ)Iң%,mrއpi/:M!&! UcS`)>D_6wn-`:!'ثH[Mb\0Yϗʘ J&=47R#i44cNva]}eˊ]%h>~vc~{,A9*&FGQ2$?6?s Iߪ((rD񰹢ɹM4bILҠ,T"MЭȯ1"PYyS ]X4~^ mIP ,gLv3"h Ƃj $-B!'ijQzɦhBL/xWtV̞P@m/DhM -0B&]2OG-C6l ENخɆPR."J`aP~$1UWؗEZEP`c[҃y^gd}g*'6[J?PHR#K1:C}Z:9]0.О.x-OR=F|2lOu'rG֡{3Qna__a4 bTIv 8=Ɲ,OWbJt$W^X}Ծ~eSsemeij6  tUIL[ЅnMRxN>v :t\'V3frGՑQbԎl%oWԁH ``bYQ [./BTu+Ε0¢o. F`Oi9)בUwZ$6gK/Q(ƀa3kTm c)t1AiuEQ RaLt8y7oW3㑳ԫ Yh<'B-SS&㧈ĢMDɸIμ73#4GJ18Y"0CyoQR~W.\+,ҶJ"ojuf\ssI '[6fF(7S@t-(":%{ ħaT6kM&r+b̷ R(a%rDi|G(J7pbK&&~Sxnf#$B)>WkB2$$g#b(G4j'tsag6p : 7I- hB;$?O5Uq;]1~7FDAD1 aV/,K,W" yz3ʆf!l6ɉ \>msF5 Z,@J_"wqZcEG9NXu(PegLUƴ]WAkF,_8fINcWں4$V;T|37Uu|zY.Wˏ6mlU&0kv᱊iV'g\ ](f".C./`X$v9-pkQħ B%C!4ڸqo#D^,3>|>"{`H(MMnEERǕs FRfh윛T;CiҷYNU_B-t|#nU%XCs?' %99},<}uMbTbi䔚^vގ$ vԎs;C6T<˅;!q$(>Y!T!2qܚTx}&#ҔI̻۵"K@YQD}~y(X(cobڟJ vI V341;!BD埓Zvbsri˟I7.Qi0t C]s&%:n{(挎l>'pF-[M"9Ǩ 7MN,wuLnUQ )2Qj&=YRλa >T8Z1J , R>w a WfK0~am ,7C^]FUF,ӂ"jc>xSG-&q޽^aTu SQ#ɖنKVQo+Ӗs;d諑j1EnʓxHd.@G{ؑ*J-{׳2:,{'4PKt+ݧϦ$k (04n2O>vgP6! dX4{!άyKGCy}PbqsMe|;EBDah?zK&+MmXY7y&k Phޤ@׹IIeR',*6H4W&lS!mnln8vW˃  \lصMjH -{i25W1Y2unRz;:l%rE3Yg׀IPJ,|R吟 o~hoP7ɗ )8Ff9ZIM_Gd=Q$i_PNrf9=C_Wt_sIn/x_(>n{5H]¿0Zznʄ9Z`&8<2zU~75hb&H,! (7%@`5hzrU*COR}9i6WǚI0EO w+X YRxϒb{zK#B{'hqM;\%txh%>&٠Ӯu>y;t,>*Q!8_hYsTFl.]>jS`kZ[ώv@@Z3 VuOr;èveg3N6fXnYst/v4}RD9`7]^3N FblϘ8X$ x}bK6ej@&;ODӼ!2m# qF6 rIrvQwȑ:lsnwz)~+yr'65#;ԆveY }x9 `ilt^@Qn~'s10.] \ :P4#xJ~w?S0o Зjh7ˋ?tƹjfqo"F `0r,5IspVJo1$ݍ sEa_B0-&o  K6>$cDm _Yfmq^,˴xz2dk0GcW CQZNA XCDf)DY"$Lh".~vjq;lGse)dz}f O8cf cn(L7@̢]7{1UWAUfHV~,$'VȽvg'1ϊpԤ` +IU[ XKq~?$48B@5f:d{_t޹wX@XúD#[B*Fbnܓ1lu*tO,c&K )$kqF4j{xEcNƹ=~CbEo dO&/G,1_{hޮTfViVīʐ#lq( p^xuxsA8x_l@B  n# F7;r2 {3>spIN`L 菱%$k:H ¡Lh0@J勢a, 7"Z$9α5+ɨ=xȝ{3LWI? \#("ȏU8V%3|*IE 3겼012Ëji~K"kI9BERN?yE$ LjV1Hy<oNNSl:O:t;n Lѫ +F6L̤+;1VI R[1$%{pOb\oXqZ\^J4q\;l z!UlX%L8"WG1̰Jb1y!q^O0K "-yꇂ8uJߠ2 ~2Q01H X‘d #luѹH϶%.68 C*Sn=ៜ9r+9TjT6Kc\}ԫ1{sG0TN n~N&0QY0vWszr"*{=8 Z85lp7!r=&^zɄHVpJj'@_Di 2{%>CC<@h.b#&ί e=;>"?>VAov7֢W  xff4`KFQ\hVl}zw)&? ǔ[#ߤ||Xeyt?p]U o^.}h4^S2oJBKy i@],x E2 4`q\"f"CCkVau@ef`#C yw!cĉ>'g0wwM'5E<[ l&AA (N1u-xH|Kp_MhK3ψw-2\̜uqAGINLh<Їڷɘ!, T@cD2Ѫpu'h?Y"R|do5s{(Xf&@u f}q'-Di(vuSDhf#Hf9)=[~@'1UY5dL90WƖZxS](<5?͍Џ>`$lˊv R$ <'7'ic4/xbO-at<;?0jml9zC.Ei|~J%-v4GF,0ͲԞxK!9LΓ>[ŕ2 'jOz X0^3g'ǁ;y$D~y:V>Z[4=ʙ e޷H䢪8ԸDJԘj ,YClsQL2;Np,X_,$>/')ThKYݦ}^FbLߪ+)z9)YkE\ ,ؽ7ٺR@r@+3/18O:?LK"\ٸ$G t},yc!ڕaTIVl8HxXi#cŔ&E߿g܉',Z\B , UR͖CF^kyCd韮>U)'"r)7A r:ZugzC8Y3^T^-ÿK!FĈVaXu8E0&.^f^h9Xwr r5zN-dk2e:Z"0  tIyU댪gIfu~ : ( 7M LK[}>v-9y f5=6:~ ~D̑O$]JځlZ"!r[mr&Ak5 c7 [a薸NII Ȼ^)?/^<֮ڐHЁ{\ɒg\) 5S.QuL61iItǰJ?E `7Xd𡆝Ba3mS gZ]-:ԛdUf6~m$IVK&L611n2Ȣt63"e%߅zÁJuZ2p~\uт5a_{pΏ|!Yt̊] g[Z5QY}M\uyՅ5Õ}1o~-7\;p$RnM+ʤDJX 桄0 +U`Sn(.+^$q~rS8=䧒ɱ^O@Z Ց#MmQD M74"dMZ3+}Y]@>cR1nRTi?IC-ɍT=N PdO{?K`؈ymMX1Ӟ!?Ch^)>8V͛Kls3'BĝĎ0Եv'KdcdfcQ |&mzpUt4c[r P؂{-s㡐f69'7j5,4o-$yFW.c|^a _tsRIe9NND;f3ָrIĄ$ЂwVky( ԫkxɡ(X{5)!J߽ $>7S8khEjNJK h^ h"I(VPYXc˱y-5?EeG:W'>L+Tu' Hay} SFȴv(Nk($rj=%JN@RT&sX`c }eB@$ꬣSD¿Hp 6::@_#x!Mt&Y?F6ww5Do#&&;iiR\&!I>U:Iʬ׋&Y@ ِp8rEp2FzwqiQ$5SV1w G-Hxeb1yUrSG͕ , P)l^/D//o[]Q䬚Wb֡ ."qib.eTN숄cg+,pw }HW($ޮ6Fr2>(h6SHzwQ%dێߘبl3;.PG?0ػT<\Ddo]4M&Aյ% 6Z72b˴w` )-A7׌o2Ÿ3btu_ *j b",XG\Di7SR7IiX1cp^+A"BeZڇmKM9;#rYmVlprL˼l'%N3ɒ[\ czRF< }5`1Q- p#1;8WT[( "9*5osPX|g} '"\ `-`{lgcVeyb׹Ij.2K2L:E#gY 69`@c)mH5IdGvKԶ?kiޚljiO8򥍮osj?Zp]/FH$)-tKd$T giV`\wɕe0Ih–e(r].֕R\!pv5oe >d J YI;j1\$[ϪPٰ 05 nﵶ{h0 I^lߺXT~WS,Ւq+NFYȢ5v.NcxV!).c62v?!_O$D\1Qgms-ҏvH "H[d$q^ozw&M[:܍_HDhK9=j6E@6.lQ ̓G >{hfITc5Vo}i6\Nesf's 4Ϣ]s30/[ >X 1`S ~TŠrAá5yN{I H02M H*IBN=oZsݗ/M?O&Hu"& ~j^1`N@.^sakILr J$i{Hʙ<6wpZ~lR"nm62|(p㹱BrBcC39 f}+@&CV7v's C%,{w[l"E`-[<to{\OqkL[nDOIENDB`@@@ NormalCJ_HaJmH sH tH R`R  Heading 1dd@&[$\$5CJ0KH$\aJ0N`"N  Heading 2dd@&[$\$5CJ$\aJ$N`RN  Heading 5dd@&[$\$5CJ\aJDA@D Default Paragraph FontRi@R  Table Normal4 l4a (k@(No List2o2 datedd[$\$B^`B  Normal (Web)dd[$\$"o" chead<o"<  pullquotedd[$\$4U`14  Hyperlink >*ph>oB>  endarticledd[$\$*W`Q* Strong5\4 @b4 Footer  !.)@q.  Page NumberH@H  5 Balloon TextCJOJQJ^JaJ\t  z z z z z z z z z (k1=J#W\ t{)( b q %u1 """ $%(9),!.k1137:8X89#=H@|CFH*HJKMMwOOSSU#WKWYZ[\\\\!Y!!!! !!! !!!!! !!!!! !!#!!!!!! !!v2!! !!! !!!! ! ! ! !!!!!!!!! !!!!!!!!!!y! b q %u1 """ $%(9),!.k1137:8X89#=H@|CFH*HJKMMwOOSSU#WKWYZ[\\\\\\\\\\\\\\\\\0000000000H0000H0H00000000H000000000000H0000000H00H00H00H00H000H000H000I0I0I0I0<I0I0X@0I0@0I0I000I0 XNC\ b q %u1 """ $%(9),!.k1137:8X89#=H@|CFH*HJKMMwOOSSU#WKWYZ[\\@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0@0 00\ $$$'[Zd37*Wdd4689d5 -/(5)7)A***677I3J5JcMMMQRZRRRRRSTS\XXCCXXXXXXX  '!! foot1upfoot2upfoot3upfoot4upfoot5upfoot6upfoot7upfoot8upfoot9upA*6IcMQZRR\*76JM?RRUS\ }y}yT}y }yT}y }y}y T}y }yԟ}y Ԡ!WWWZZZ1[1[\ !WWWZZZ7[7[\ i*urn:schemas-microsoft-com:office:smarttagsState0http://www.5iamas-microsoft-com:office:smarttagsh*urn:schemas-microsoft-com:office:smarttagsCity0http://www.5iamas-microsoft-com:office:smarttagsZ *urn:schemas-microsoft-com:office:smarttags PlaceTypehttp://www.5iantlavalamp.com/Z *urn:schemas-microsoft-com:office:smarttags PlaceNamehttp://www.5iantlavalamp.com/_*urn:schemas-microsoft-com:office:smarttagscountry-regionhttp://www.5iantlavalamp.com/V *urn:schemas-microsoft-com:office:smarttagsplacehttp://www.5iantlavalamp.com/ p7"     33!6'6 7&7 [[[[\\..FFQQ[[[[["\\\::::::MMwOOSS#WKWZ[\\\\w/p+?w/kogGpZ+? g+?pqu+? 5@4W4W0, vv4W4W\ @UnknownG:Ax Times New Roman5Symbol3& :Cx Arial5& :[`)Tahoma"qhBLGLGL N / N /!24d\d\3QHX)?2On the origin of strategies JKUATJKUATOh+'0  ( H T ` lx On the origin of strategies JKUATNormalJKUAT2Microsoft Office Word@^в@&@<'ͭ&@&  N՜.+,D՜.+,L hp  JKUAT/d\ On the origin of strategies Title0 8@ _PID_HLINKSA62Chttps://www.mckinseyquarterly.com/On_the_origin_of_strategies_1060foot92Chttps://www.mckinseyquarterly.com/On_the_origin_of_strategies_1060foot82Chttps://www.mckinseyquarterly.com/On_the_origin_of_strategies_1060foot72Chttps://www.mckinseyquarterly.com/On_the_origin_of_strategies_1060foot62Chttps://www.mckinseyquarterly.com/On_the_origin_of_strategies_1060foot52Chttps://www.mckinseyquarterly.com/On_the_origin_of_strategies_1060foot42 Chttps://www.mckinseyquarterly.com/On_the_origin_of_strategies_1060foot32Chttps://www.mckinseyquarterly.com/On_the_origin_of_strategies_1060foot22Chttps://www.mckinseyquarterly.com/On_the_origin_of_strategies_1060foot1  !"#$%&'()*+,-./0123456789:<=>?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\]^_`abcdefghijklmnopqrstuvwxyz{}~Root Entry F` &Data ;܁1Table|y(WordDocument.tSummaryInformation(DocumentSummaryInformation8CompObjq  FMicrosoft Office Word Document MSWordDocWord.Document.89q