Successful Share-Building Strategies

Market share helps determine business performance. Not only do marketing executives say so but abundant evidence supports the statement. Past data analyses have consistently demonstrated that most high-share businesses enjoy above-average profit margins and rates of return on investment, while most small-share businesses have below-average margins and ROI.1 This is true even when allowance is made for other factors that typically accompany high market share, such as above-average product quality, early entry into the market, and broad product lines. Previous research suggesting the relationship between market share and profitability came from the Profit Impact of Marketing Strategies (PIMS) program sponsored by the Strategic Planning Institute in Cambridge, Massachusetts. Since analyses of those results were first published in the mid-l970s, the PIMS data base has been greatly expanded. Information is now available for a much larger number of businesses—about 2,000 as of early 1980. The data also cover a longer time span, ranging up to nine years for some of the participants. Moreover, a wider variety of companies and industries are now represented in the data base, including smaller companies and a significant number of corporations based outside North America. (For further details of the PIMS program, see the sidebar.) The PIMS Program The statistical analyses summarized in the exhibits used the data base of the Profit Impact of Market Strategies (PIMS) program. PIMS is an ongoing program of research and analysis on business strategy conducted by the Strategic Planning Institute in Cambridge, Massachusetts. As of early 1980, more than 200 corporations were participating in the program. Each company submits financial, market, and competitive information for a number of its divisions and products and receives a variety of diagnostic reports on each business. These reports provide guidelines for explaining current performance and for developing future business strategies. All of the PIMS data pertain to companies’ business divisions, product lines, or other subdivisions within companies that can be distinguished from other parts of companies in terms of several criteria. Ideally, each business unit serves a distinct, well-defined market; has a clearly defined set of competitors; is largely self-sufficient in terms of technology, operations, and marketing; and has, or could have, its own business strategy. Following these guidelines, the managers and planning specialists of the companies participating in the PIMS program develop their own definitions of businesses, estimates of market size and growth rate, and allocate their own accounting data among businesses as appropriate. The products and industries are not identified in the data base, and all financial data are disguised by applying a common scaling factor to actual monetary amounts. While the PIMS data base has changed considerably since the program’s inception in 1972, the relationship between market share and profitability has not. Exhibit I shows average pretax rates of ROI for businesses grouped according to increasing market share, based on four-year periods. Exhibit I Market Share and Return on Investment The exhibit shows ROI increasing steadily as market share rises. The average rate of return for business units with shares of more than 40 % is two-and-a-half times the average for those with shares of 10 % or less. Put another way, a difference of 10 points in market share is accompanied, on average, by an increment of about 5 percentage points in ROI. Within each group, profit variations reflect differences in other factors related to ROI, such as capital intensity, vertical integration, and market growth. Altogether the PIMS statistical models include 28 factors related to profitability, of which market share is only 1. But when all of these effects are taken into account, the net influence of market share remains strong. The financial advantages of a strong market position make understandable the fact that a common strategic goal is to increase market share. Only market leaders, such as General Motors and IBM, are satisfied just to hold on to their existing market shares. In most markets, the number two and three competitors—the Fords, Millers, and Schlitzes—are aggressively trying to catch the leaders. Except perhaps in embryonic or rapidly expanding market environments, competitors ranking lower than fourth or fifth must increase share. The alternatives are unpalatable: to hang on with an unsatisfactory rate of ROI, often below the cost of capital, or to liquidate.2 Thus, the vast majority of businesses would like to increase their market share, and in all likelihood most are actively trying to do so. What are the most promising means? In this article, we present some approaches for increasing market share based on two types of information. First, we have analyzed the market share changes of more than 1,200 businesses participating in the PIMS program. Second, we have reviewed case studies, corporate annual reports, and other published accounts of the competitive strategies used to build market share in a wide variety of industries and product categories. These descriptions add a qualitative dimension to the standardized financial and market data available from the PIMS program. Developing a Market Share Model Market share can be affected by many different aspects of competitive strategy. The concept of the marketing mix is based on the working together of all dimensions of marketing strategy—product policy, pricing, distribution, sales force efforts, advertising, and promotion—to influence buyer choice. Not only do management’s own policies and programs affect sales, but so do those of competitors. A realistic approach to explaining market share change must, therefore, include all of the important elements of marketing strategy and must somehow relate any single competitor’s actions to the contending moves of rivals. Since computers first came into general use in marketing research during the 1960s, numerous efforts have been made to develop statistical models to explain and predict changes in market share. Quite a few of these models have been described in published accounts; no doubt even more remain in companies’ confidential files. With few exceptions, models of market share change are designed for single product categories such as beer, cigarettes, or television sets, and the results are intended to be used only by marketers of the product studied. Our analysis of the PIMS data base employed a very different approach: we related market share changes to competitive strategies for a heterogeneous cross section of businesses. We divided the overall sample into three subgroups—consumer products, raw and semifinished materials, and manufactured industrial products (capital goods, components, and supplies)—in the belief that major differences would exist among the three in the importance of such strategic factors as pricing and advertising. No doubt the marketing manager of a producer of ball bearings would prefer to base decisions on a model of market share change specifically tailored to that industry. But the data required to develop statistical models are simply not available for many products. Moreover, one-at-a-time studies of individual markets do not provide a basis for reaching any general conclusions about how market share can be gained or lost—unless, of course, hundreds of similar studies can be assembled and compared. Pending this unlikely occurrence, findings derived from a diverse cross-sectional data base can provide valuable insights into the common denominators of successful share-building strategies. Measuring Competitive Strategies For each of the businesses in the PIMS data base, annual information is available on many of the important elements of competitive strategy. These include: Relative product quality. This is measured through an index of the quality of each business unit’s products or services relative to those of its major competitors (derived from judgmental estimates supplied by line executives and/or planning staff). The judgmental assessments indicate, on an annual basis, what proportion of total sales of products or services was accounted for by each of three categories: those superior to competitors’, those of equivalent quality, and those of inferior quality. The index used in our analysis (and in other PIMS research) is computed by subtracting the percentage of inferior quality sales from the percentage of superior quality sales. Relative new products. This is the proportion of a business unit’s sales accounted for by new products or services—again, relative to major competitors. New products are defined as those introduced during the preceding three years. If a business with 30 % of sales in new products were competing with companies estimated to have 20 % of new product sales, our index would be 30 % minus 20 % , or 10 % . Marketing expenditures. Current outlays for marketing are subdivided into three categories: sales force, media advertising, and sales promotion. Sales promotion costs include contests, premiums, trade shows, distributor and dealer incentives, temporary price reductions, and free goods. To estimate the relationship between marketing effort and market share change, we calculated the percentage change in each type of expenditure and adjusted it for overall market growth. (The nature of these adjustments is explained in a later section.) Pricing. Each business unit supplied estimates of its prices relative to competitors’, expressed in index form. If prices are equal to competition, the index is set at zero; a 5 % premium is expressed as +5, a 5 % discount as –5, and so on. Measuring Market Share Change Market share can be defined and measured in several ways.3 When all of the competitors in a market sell a single type of product, like gasoline or soft drinks, their shares of total volume can be calculated in physical units. Most of the business units represented in the PIMS data base, however, produce and market several different types of products or services, as do their competitors. Consequently, the only practical way to measure market share is in terms of monetary value. Each business unit’s market share is simply its dollar sales divided by estimated total sales in its served market. The concept of served market also deserves some comment. How should a business define its market for purposes of measuring and tracking market share? Usually more than one definition is possible.4 Consider, for example, cooking appliances. There are electric ranges, gas ranges, countertop microwave ovens, combination microwave-conventional ranges, and convection ovens. From the viewpoint of, say, Amana, should all of these types of cooking appliances be considered a single market? Or, at the other extreme, should each category be considered a separate market? Also, in view of the substantial quantity of microwave oven imports from Japan, can the United States realistically be viewed as a market unto itself, or is the scope of the market worldwide? There are no clear-cut answers. In the PIMS program, each business develops its own definition of the served market. The scope of the served market is limited in terms of products, customers, and geography to those sectors or segments within which a business unit actively competes. The market share data reported by the PIMS participants are based on this concept. The method we have used to measure change in market share also requires a brief explanation. We measure change in terms of average annual percentage rate of increase or decrease in market share. For example, a change of +10 % could reflect either of the following situations: We measure change in market share this way because all our measures of competitive strategy are defined in terms relative both to each business unit’s size and also to competitors’ activities. For example, our index of new product activity is the percentage of sales accounted for by new products minus the estimated ratio of new products to total sales for major competitors. Suppose that a business has 20 % of its sales in new products. The absolute amount of new product activity that this represents in a given market will depend on the size of the business: it would be twice as great for a business with a 50 % market share as for a competitor with a 25 % share. Similarly, relative product quality is measured in terms of percentages of a business unit’s total sales, and changes in marketing expenditures are measured relative to previous levels of expense, which clearly depend on sales size. Thus, the amount of effort required to achieve a given increase in market share—in terms of these “relative change” measures—depends on one’s starting point. If, for example, a 10 % improvement in relative quality on the part of a business with a 50 % market share leads to a gain of two points, we would expect the same relative improvement by a competitor with a 25 % market share to yield a gain of one point. We carried out our analysis, then, in terms of percentage rates of change in market share. For each business, an average annual rate of change over a four-year period was calculated by fitting a trend line to the actual year-to-year changes. Patterns of Change Using the measures of competitive strategy and market share change just described, we employed a variety of statistical methods to analyze patterns of change among the businesses in the PIMS data base. (While the results are presented in this article in the form of tabulations, our principal method of analysis was multiple regression, in which all of the factors discussed here—and a few others—were simultaneously related to change in market share.) We found that the strategic factors generally involved in market share gains include all of the following: Increase in new product activity. Increase in relative product quality. Increase in expenditures for sales force, advertising, and sales promotion, relative to the growth rate of the served market. Our analysis also showed that changes in a business unit’s market share depend on its competitive position within its market. Specifically, as shown in Exhibit II, changes in share are related to both market share and market share rank. Because all three types of businesses—consumer products, raw materials, and industrial products—displayed the same basic pattern, they have been combined. The data indicate that businesses with market shares below a certain critical level tend to gain share, while those with shares above that level tend to lose. Exhibit II Changes in Market Share Related to Beginning Market Share and Competitive Rank A Possible Explanation Why are changes in market share so strongly related to beginning share level and rank? In a similar study of the PIMS data, we found that the relative sizes of competitors in a market tend to follow a common basic pattern. If the second-ranking competitor is two-thirds as large as the leader, then the third-ranking competitor is about two-thirds as large as the second, and so on down the line.5 The size ratio that best fits the actual distribution of market share varies from one market to another, but the values for most markets cluster around the overall average of 0.63. The critical levels of market share for each competitive rank are the values 40 % , 25 % , 15 % , and 10 % , which form a progression in which each is between 60 % and 65 % as large as the next highest value. In a given market, a natural equilibrium exists in terms of the relative sizes of competing businesses. For most markets, a good approximation of the natural structure is one in which the market leader has a 40 % share, and numbers 2, 3, and 4 have shares of 25 % , 15 % , and 10 % , respectively. Apart from changes in position caused by major shifts in competitive strategies, each competitor will tend to gain or lose share according to whether it is below or above the natural level. This may appear to be a rather abstract theory for such a practical matter as change in market share. Most marketing executives probably believe that gains in share must be earned and can be explained sensibly only in terms of tangible advantages in product features, customer service, price, or other specifics. Why should there be any natural tendencies for share to rise or fall? At present we cannot give a complete explanation of the pattern shown in Exhibit II, but we are convinced that it is real. While we are not aware of any theory which would have led us to expect such a pattern, we believe that the way changes in share relate to share level and competitive rank reflects powerful forces operating in most markets. Most of the businesses in the data base enjoyed gains in market share. How can this be? If one competitor gains, must not another lose? Clearly, the PIMS data base does not contain a proportional representation of losers. One reason is that the small-share businesses included only survivors in each marketplace. Some of their gains came at the expense of declining or discontinued businesses, which are less likely to be involved in the PIMS program than those with stable or improving positions. Influence of New Products Developing and introducing new products is one important and widely used approach to building market share. Exhibit III shows average percentage changes in market share for consumer, raw material, and industrial products businesses as a function of varying levels of new product sales and changes in new product activity. (The data in Exhibits III, IV, and V are based on a sample that excludes businesses which sold most of their output to other units within the same parent corporation. Also excluded are businesses whose ability to gain share was restricted by limitations in plant capacity.) Exhibit III Changes in Market Share and New Product Activity Levels In general, these data indicate that gains in market share are associated with either a high level of new product activity (relative to competitors) or an increase in new products, or both. Some caution is required in interpreting the pattern shown in Exhibit III because the two variables—levels and changes in new products—are inversely related. For example, industrial product manufacturers that reduced new products relative to competitors gained market share more rapidly (+3.4 % ) than those that increased their new product activity (+2.9 % ). But most of the former group started with high proportions of new products to total sales. When we consider both factors in combination, along with all of the other measures of competitive actions, each is significantly related to market share change. Product innovation is universally recognized as a strategy for building market share—in both mature and expanding markets. Marketers of computers and semiconductors are continually introducing newer, more powerful, more versatile models. Producers of processed food and personal care and household products regularly offer “new, improved” formulations, flavors, sizes, and package variations. Even traditionally staid AT&T, faced with direct competition for the first time in the 1970s, sought to regain lost market share with a series of new types of telephones and other products.6 Influence of Product Quality A strategy of building market share via new products frequently involves improvements in the quality of product or service offerings. Sometimes the quality of established products is upgraded by a more gradual process: automobiles are made more fuel efficient, household appliances more compact, industrial machinery simpler to operate. Whether connected with new product introductions or not, quality improvement is a powerful means of building market share, as shown in Exhibit IV. In all three groups of businesses—consumer products, raw materials, and industrial products—competitors that increased relative quality enjoyed much greater gains in share than those whose quality ratings remained constant or diminished. Exhibit IV Changes in Market Share and Quality Improvement Another indication of the importance of product quality is that among business units achieving substantial market share gains (5 % or better annual increases), nearly half reported at least moderate improvements in relative quality. Some recent competitive successes stemming from qualitative factors have been accomplished by Japanese producers of automobiles, cameras, and electronics. A study comparing the U.S. and Japanese color television industries, for example, showed that Japanese-made sets had much lower failure rates. One measure of the difference was that the number of defects per 100 sets shipped by a Motorola plant fell from an average of 165 to 3 or 4 within three years after the plant came under Japanese management.7 Small wonder, then, that Japanese television producers have enjoyed steady market share gains. A strategy of building share via quality improvement does not necessarily imply offering deluxe products. We are not suggesting, for example, that Chrysler rebuild its position in the automobile market by emphasizing models more luxurious than Mercedes or Cadillac. Quality, like beauty, is relative, and, in most markets, offering better value—especially in the moderate-priced sector where volume is usually concentrated—is most important. The spectacular growth of Bic in the ball-point pen market, beginning in 1958, illustrates the idea. Bic succeeded by providing much better quality at 19 cents than others did at similar or higher prices-not by marketing fancy pens at $ 5. (Cross pens are also successful and highly profitable, of course, but on a smaller scale.) Influence of Marketing Budgets A third major element of share-building strategy is the marketing budget—the amounts spent for sales force, advertising, and sales promotion. Ideally, each category of promotional outlay should be measured relative to spending by competitors. We were unable to do this directly because estimates of competitive expenditures were not available. Instead, we related changes in each budget category to the concurrent rate of growth or decline in the size of the served market. Specifically, we first calculated percentage rates of change in each category of expenditures; for example, a business that increased sales force spending from $ 1 million in one year to $ 1.2 million in the next would be treated as a 20 % increase. Next, we computed an average annual rate of change for the entire four-year period and also the rate of change in market share. The final step was to adjust each business unit’s change in marketing expenditures by the rate of growth or decline in its served market. We observed, however, that marketing budgets tend to lag behind gains in sales. We thus analyzed the data base to determine the routine rate of change in each expense category for a given rate of overall market growth. Based on just those businesses whose market share remained essentially unchanged, we found that for each 1 % change in the size of the served market, spending on marketing activities changed on average by the following amounts: (We estimated these relationships by a regression analysis in which we used the rate of growth in the served market to predict the rate of growth in each category of expenses. We restricted the sample to businesses with market share gains or losses of less than 5 % . Because of this, changes in the size of the market were essentially equal to changes in sales volume for these businesses.) Such typical relationships between volume change and marketing budget change are of some interest in and of themselves. For one thing, all categories of marketing expense tend to change much less rapidly than sales volume—typically about half as much. The implication is that, with long-term growth in most markets, average ratios of marketing costs to sales should decline. A reason for marketing budgets lagging behind sales growth during the 1970s may have been that rising raw materials costs forced selling prices up faster than the costs of such marketing resources as salespeople’s salaries and advertising media. In addition, we suspect that a degree of inertia often interferes with upward adjustment of budgets as sales volume expands, particularly in fast-growing markets. Marketing budgets are apparently most closely tied to sales variations for consumer products companies and least so for raw materials producers. Indeed, among the latter group only sales force spending displays any systematic relationship to change in volume. We used the previous relationships to calculate adjusted rates of change in marketing spending. For example, suppose a consumer product business increased its sales force outlays by 15 % annually with its market growing by 10 % . The normal rate of change would be calculated as follows: 10 % market growth multiplied by 0.6, or 6 % , and the adjusted rate of increase for the business is 15 % (actual) minus 6 % (normal), or 9 % . We then related these adjusted rates of change to change in market share. We reasoned that increases greater than normal, based on market growth, usually represented gains in a business unit’s share of total marketing effort within its market, while less than normal increases implied spending reductions relative to competition. Exhibit V summarizes the relationships between market share change and adjusted rates of change in marketing expenditures. Clearly, all three forms of promotional effort—personal selling, advertising, and sales promotion—can build market share. Also important, the impact of each type of spending varies considerably among the three types of businesses. Exhibit V Changes in Market Share and Changes in Spending for Sales Force, Advertising, and Sales Promotion Of all the competitive strategy elements we analyzed, sales force expenditures are most strongly related to market share changes. Intensified selling effort appears to be a common feature of share-building strategies for both consumer and industrial products; its involvement is much less clear for producers of raw materials. Advertising outlays are significant contributors to improvements in market share only for consumer products. Among marketers of industrial products and raw materials, advertising is typically a minor component of the overall marketing budget and presumably has a correspondingly small role as a competitive weapon. (There are, of course, exceptions: advertising has apparently been used effectively by some manufacturers of office typewriters, copying machines, and farm equipment.) Interestingly, changes in sales promotion spending are reasonably important in explaining market share changes for all three types of business. Because the sales promotion category includes a wide variety of competitive activities, the ways in which it is used undoubtedly vary a great deal. For consumer products, temporary price reductions, free samples, and display allowances are widely used forms of promotion, especially in connection with new products. In recent years, durable goods manufacturers have also extensively employed cash rebates. Industrial goods producers use trade shows and exhibits, catalogs, distributor incentive programs, and temporary price reductions as promotional devices. Efforts to relate changes in sales or market share to marketing expenditures have always been plagued by problems of separating cause from effect. The figures shown in Exhibit V are no exception. For example, the data suggest that industrial goods producers which increased sales force budgets by 5 % (after adjusting for market growth) enjoyed substantially greater gains in market share than if they had kept their spending in line with the overall market. Did higher outlays on sales force activity cause market share gains, or were budgets expanded in line with growth in sales emanating from other sources? Our analysis does not permit a clear-cut answer. All that we can say with certainty is that increased outlays on all three categories of marketing expense—and especially on sales force activity—tend to accompany successful share-building strategies. No doubt higher spending levels are usually both causes and effects of sales expansion. Noninfluence of Pricing One element of competitive strategy that we found unrelated to change in market share was businesses’ price levels relative to competitors. This finding (or nonfinding) will surprise many, especially in view of the emphasis placed on aggressive price-reduction strategies in recent publications on strategic management. An often-cited example is that of Texas Instruments in the hand-held calculator market in the early 1970s. TI supposedly cut prices in anticipation of cost reductions achieved through a substantial advantage in accumulated volume (or experience). In much the same way, Japanese producers have established strong positions in U.S. and world markets for motorcycles, cameras, television sets, and audio equipment, among others, partly by pricing at lower levels than established competitors. For that matter, Henry Ford used the same approach to achieve leadership in the automobile industry in the early twentieth century. These examples illustrate the point that strategies of price reduction are used, sometimes with dramatic effects, to build market share. If this is true, why did we find no relationship between price and market share among the businesses in the PIMS data base? The explanation appears to be that growth via rapid price cutting is not common in mature markets. Most of the PIMS businesses operated in relatively mature competitive environments rather than embryonic industries such as videocassette recorders or electronic word processing systems. Perhaps opportunities to gain volume through aggressive price reduction are rare when technology and marketing practices are well established. Moreover, price reductions in mature markets by one competitor are likely to be met by rivals, making it more difficult to achieve any advantage in relative price. Whatever the reasons, major changes in price relative to competitors were not common among the businesses in the PIMS data base: fewer than 10 % of them either raised or lowered relative prices by as much as 3 % per year. Possible Share-Building Strategies The experiences of the businesses in the PIMS data base, as summarized in Exhibits II–V, provide useful clues for managers and planners seeking ways to build market share. To recap, the PIMS data show that gains in market share are associated with product quality improvements, new product introductions, and marketing budget increases. Our analysis also showed that, at least within our sample, price reductions are seldom used as a means of building share. Also, we found that business units tend to gain or lose share depending on whether their market shares are above or below a critical level, which in turn depends on the unit’s competitive rank. For example, market leaders tend to lose share if they hold more than 40 % of the served market and to gain share if they hold below 40 % . Taken together, the effects of the strategic factors depicted in Exhibits II–V, along with the impact of competitors entering or leaving markets, account for much of the shifting in shares that took place in the markets in which the PIMS businesses participated. We developed statistical models that simultaneously incorporate all of these factors. These models explain about 40 % of all the variation in market share changes for consumer products makers and about 30 % of the variation for both raw materials and industrial products manufacturers. While it has provided some clear-cut results concerning factors that affect market share, our analysis failed to isolate certain other forces. Some specific ways of competing are distinctive to particular industries or products and cannot readily be measured and analyzed in a multi-industry program such as PIMS. No doubt, much of the shifting in market position not attributable to changes in quality, new products, or marketing budgets was due to the use of these industry-specific competitive tools. In addition, enormous differences must exist in the quality of execution of competitive strategies. An increase of 20 % in the sales force budget is likely to be much more effective for a business that has well-trained and well-managed salespeople than for a business where the efforts are just as expensive but less adequately executed. It is not surprising, therefore, that our analysis did not explain all of the changes in market share that took place in our sample. Beyond Statistical Considerations Based on our review of case studies and other published accounts of share-building strategies, we can add two more guidelines to those derived from a purely statistical analysis: 1. Successful share-building strategies usually involve a combination of several competitive factors. The data in Exhibits III–V show how quality improvements, new products, and marketing expenditures are related to change in market share. This might be interpreted to mean that any one of these elements of competitive strategy could be used independently of others. While situations probably exist in which a single competitive weapon, such as advertising, can be employed to build share without changing anything else, we think they are uncommon. Typically, better results can be obtained by using a balanced, consistent marketing program or a mix of strategic factors. In fact, among the businesses in the PIMS data base that achieved significant gains in market share, only 17 % changed just one of the five strategic factors shown in Exhibits III–V. Most employed two or three factors in combination. For example, quality improvements and new product introductions were almost always accompanied by increases in at least one of the three categories of promotional spending. This pattern is consistent with the notion of the marketing mix, which assumes that all elements of a marketing program are mutually consistent. If any one of these factors changes, it is reasonable that others would need to be changed accordingly. A classic illustration of the effectiveness of a balanced marketing program is evident in the experience of L’eggs pantyhose.8 The Hanes Corporation introduced L’eggs in 1971 with a marketing strategy that included several novel elements: a one-size product to fit most users, a new type of package, a special display fixture, a new system of direct-to-the-store distribution, and heavy introductory advertising and promotion. By 1974, L’eggs was the leading brand in the pantyhose market. While each component of the strategy undoubtedly contributed to the product’s success, it also seems clear that the components reinforced each other. For example, heavy advertising and promotion speeded up consumer trial; that facilitated acceptance by retailers; and the system of direct distribution ensured that L’eggs would seldom be out of stock, allowing satisfied buyers to develop steady repeat buying routines. The value of a balanced marketing program, in which each of several elements plays a part, is hardly a novel idea. But it does merit repeating and is underscored by the experience of most of the PIMS businesses that achieved share gains. 2. Most successful share-building strategies are based on giving primary emphasis to one or more segments within the served market. The PIMS data do not break down market volume into segments or submarkets. Yet most markets include several—sometimes many—distinct subdivisions of customer groups, product variations, price versus quality levels, or distribution methods. We are confident that if data on segments within markets were available, they would show that most of the businesses achieving major share gains did so by focusing their efforts on selected segments—often ones that were relatively small at first. Some of the most dramatic and well-publicized strategic successes of recent years have been based on the principle of segment focus that we advocate. For example, Philip Morris steadily improved its share of the cigarette market throughout the 1970s. Part of the company’s success has been due to its heavy support for Merit, a low-tar brand that apparently had special appeal for health-conscious smokers. With similar heavy support, Philip Morris’s Miller Beer subsidiary introduced the first successful “light” beer in the mid-1970s. Miller enjoyed substantial gains in market share, thanks in part to aggressive advertising aimed at diet-conscious drinkers. Honda entered the U.S. motorcycle market in 1959 by offering high-quality small motorcycles at a time when the established producers regarded such products as toys. Moreover, promotion for Honda was directed to an entirely new type of customer: the suburban, middle-class male. By the mid-1960s Honda’s market share was 63 % —a position achieved by discovering a new market segment rather than by attempting to compete head-on with the American and British producers that had previously dominated the industry. Focusing on specific market segments as a growth strategy is not unique to consumer goods markets. Since the mid-1970s, two small companies—Savin and Canon—have successfully challenged Xerox in the office copier market by offering relatively low-priced, small machines suitable both for small users and for decentralized copying services in large organizations. Savin’s low cost—one model was priced almost 50 % below the lowest-priced Xerox model—led many users to switch to what might be called “distributed copying” in much the same way that computer terminals and low-priced minicomputers opened the way for distributed computing.9 Whether Savin planned it that way is unclear, but a small market segment mushroomed—so much so that by 1978 Savin was supposedly selling more copiers than Xerox and IBM combined! (The company’s market share measured in dollars, however, was much smaller.) Canon followed essentially the same strategy as Savin during this period and also prospered. According to Sanford Fox, general manager of Canon’s copier division, “…we saw an unfilled niche…that was created by the strategy of the established companies. The leader—and the followers—hadn’t given much attention to ‘convenience copiers’…we based our strategy on focusing our resources on this slice of the total market.”10 We believe that most successful share-building strategies, like those of Philip Morris, Miller, Honda, Savin, and Canon, are based both on emphasizing specific market segments and designing products and marketing programs to meet the special needs or interests of these segments. Effective strategies for building market share cannot be reduced to formulas. The trick is to find a segment with sufficient potential for growth that has been neglected. To do so, we suspect, requires substantial creativity or a lot of luck—or both. 1. For summaries of earlier work, see Sidney Schoeffler, Robert D. Buzzell, and Donald F. Heany, “Impact of Strategic Planning on Profit Performance,” HBR March–April 1974, p. 137; and Robert D. Buzzell, Bradley T. Gale, and Ralph G.M. Sultan, “Market Share—A Key to Profitability,” HBR January–February 1975, p. 97. 2. A prominent consultant has suggested that a stable market never has more than three significant competitors. See “The Rule of Three and Four,” Perspectives, no. 187 (Boston: Boston Consulting Group, 1976). 3. Alfred R. Oxenfeldt, “How to Use Market-Share Measurement,” HBR January–February 1959, p. 59. 4. The problem of proper market definition has received much attention in connection with the application of strategic management concepts. For a discussion, see Robert D. Buzzell, “Note on Market Definition and Segmentation,” no. 9-579-083 (Boston: Intercollegiate Case Clearing House, 1979). 5. Robert D. Buzzell, “Are There Natural Market Structures?” working paper, Harvard Business School, October 1979. 6. “Selling Is No Longer Mickey Mouse at AT&T,” Fortune, July 17, 1978, p. 98. 7. J.M. Juran, “Japanese and Western Quality: A Contrast in Methods and Results,” Management Review, November 1978, pp. 27–28, 39–45. 8. Based on F. Stewart DeBruicker and Harvey N. Singer, “L’eggs Products, Inc.,” no. 9-575-065 (Boston: Intercollegiate Case Clearing House, 1974). 9. “Savin’s Savvy Sell,” Sales and Marketing Management, October 10, 1977, p. 34. 10. Quoted in an advertisement for the Wall Street Journal that appeared in Advertising Age, February 11, 1980, p. 15.

Successful Share-Building Strategies

Partner Center