Marketing to the Wine Consumer--An Overview

Raymond J. Folwell, Washington State University
ABSTRACT - Wine buyers were identified by major demographic characteristics, wine products purchased, and intended use of the products. The typical wine purchasing household has higher income, fewer members, and more education than nonpurchasing households. About one-half of U.S. households never buy wine, and less than 5% purchase more than one-half the wine.
[ to cite ]:
Raymond J. Folwell (1980) ,"Marketing to the Wine Consumer--An Overview", in NA - Advances in Consumer Research Volume 07, eds. Jerry C. Olson, Ann Abor, MI : Association for Consumer Research, Pages: 89-94.

Advances in Consumer Research Volume 7, 1980     Pages 89-94


Raymond J. Folwell, Washington State University

[Professor and Agricultural Economist, Department of Agricultural Economics.]


Wine buyers were identified by major demographic characteristics, wine products purchased, and intended use of the products. The typical wine purchasing household has higher income, fewer members, and more education than nonpurchasing households. About one-half of U.S. households never buy wine, and less than 5% purchase more than one-half the wine.


Wine, as a beverage, has only been recently discovered by the average U.S. consumer. Prior to the late 1960s, the annual per capita consumption of all wine was approximately 1 gallon. Then in the late 1960s and early 1970s, the "wine boom" occurred and the per capita consumption of wine increased 10% to 14% per year. Since the wine boom, the annual growth in the per capita consumption of wine has returned to a more normal sales growth rate of approximately 5% annually.

Before we initiated a national wine marketing study, the available data relating to the U.S. wine market was not adequate to make many marketing and merchandising decisions. The information that was generally available consisted of those statistics compiled by various governmental agencies because wine is a beverage which contains alcohol and is regulated at both the state and federal levels. Our national wine marketing study generated information on who buys wine in the U.S., when and for what occasions it is bought, what are the general characteristics of the wine market, and the demand functions for wine in the U.S.


As implied above, the only wine consumption related data was shipment data from secondary sources as secured from government reports on tax withdrawals. The data generated in our national wine survey was from a panel of households in the U.S., consisting of about 7,000 households at any one time between February, 1975, and January, 1976. The households were recruited on a regional basis (Figure 1).

Members of this panel reported their monthly wine purchases that were related to household use of the wine. The households did not report wines purchased for consumption away from a household setting, such as in a restaurant.

All of the major demographic features of each individual household on the panel, such as age, sex, income, occupation, education, race, marital status, and region of the country, were available. These household demographics were cross tabulated with each household's purchasing pattern so that insights into consumer buying attitudes and patterns could be developed.

As with most household panel data, the data from the panel of households we used had some limitations. The panel of households was not a probability sample.



Single member households were not included. There was some turnover in the households on the panel. Because of these limitations, the panel did not match in all instances the estimated percentage distributions of households in the U.S. according to major census household demographics. Overall, the panel over-represented middle-aged households with at least some college education and annual incomes in excess of $10,000. Complete descriptions of the household panel, purchase diary, and the volumes of wine purchased are contained in reference #1.

Demographics of Purchasing and Nonpurchasing Households

Only the more important demographics identified in previous work were used in this paper to describe wine purchasing and nonpurchasing households (2). These demographics were the age of the wife or female head of the household, age of the male head of household, education level of the male head of the household, size of family or household, and household income. The computed mean values of these major demographics for the purchasing and nonpurchasing households are in Table 1.

The wife and male head of household tended to be slightly older in the purchasing households than in the nonpurchasing households. The only significant differences at the .05 probability level between the purchasing and nonpurchasing households were the age of the male head of the household in the East South Central and South Atlantic regions, and the age of the wife in the East South Central region.

Family or household size was smaller for the wine-buying households than for the nonbuying households on a national basis. Regionally, the family sizes of purchasing households were the same or smaller for the nonpurchasing households. At the .05 probability level, the only statistically significant differences in the sizes of families between the purchasing and nonpurchasing households were in the Middle Atlantic, East South Central, Mountain, and Pacific regions.

Nationally, the male head of household of the wine-buying households had more years of education than those of nonpurchasing households. The levels of education of the male heads of households for the purchasing households in each region were significantly higher than for the nonpurchasing households on a statistical basis at the .05 probability level.



The average household incomes of the wine-buying households were significantly higher at the .05 probability level than for the nonpurchasing households. The wine purchasing households on the panel, as compared to the nonpurchasing households, can best be characterized as being slightly further along in their life cycles, having smaller families, more highly educated male heads of households, and significantly higher incomes.

The demographics of the households purchasing the various wine types were different. A tabular summarization of these demographics over all regions is shown in Table 2. Households that bought table wines had more incomes than the average household that bought other types of wine. In contrast, the households purchasing the other wine types tended to be slightly less educated, with smaller household incomes, and larger families. Thus, it can be concluded that various wine types have certain types of households that are purchasing them and they can be targeted in the marketplace.

The average size purchase was much greater for nonvarietal table wine (78.3 oz.) than any of the other wine types. Such a relationship would be expected because of the importance of nonvarietal table wine in everyday consumption patterns. The other types of wines appeared to be purchased in about the same average quantities (40-49 oz.), with brandy being the lowest with 30.8 oz.



The average number of purchases per wine purchasing household was greatest for nonvarietal wine at 3.6. Dessert wines were next in terms of average purchases per purchasing household. All the other wine types were purchased in about the same frequency.


Females accounted for about 60% of all the wine purchases. Males were not the dominant purchasers for any wine type. They were most important in buying varietal table wines. On a regional basis, there are exceptions to the above generalization that females were the dominant purchasers.

The sex of the wine purchaser was related to the place of purchase as well as the type of wine (Table 3). The supermarket was the dominant place in which the wine purchases took place. The liquor store was second in importance with all other types of retail outlets being of minor importance. There were exceptions on a regional and wine type basis because the place of purchase varies with state laws. For example, the liquor store clearly dominates the reported purchases of table wines in the Middle Atlantic region, where most states permit wine sales only in a state liquor store. In contrast, the supermarket clearly was the most important retail outlet for varietal and nonvarietal table wines in the Pacific region, where it is legal to sell such wines through supermarkets or food stores. In the Pacific region, the female purchaser of varietal table wines was 61.2% of all purchases, while in the Middle Atlantic region, the female purchaser was only 48.4% of all purchases for such a wine. It appears that the authorized retail sales agency is an important factor in explaining whether men or women buy most of the wine. However, the type of wine to be purchased also appears to be highly related to the sex of the purchaser. A plausible hypothesis is that the type of wine bought is related to its intended use and, thus, the purchaser is somewhat, although not directly, related to the type of authorized retail outlet.



The households were provided space on the monthly diary to mark whether the wine was bought for a special occasion, everyday use, cooking, or as a gift. Table 4 summarizes the percentage distributions of the intended uses by wine type for the entire United States. Everyday usage was the dominant intended use of all wine types except for sparkling wines, which were intended to be used primarily on special occasions. Cooking and wine intended as a gift were not significant uses of any of the wine types. On a regional basis, there were some exceptions.



Table 5 indicates the number of responses and percentage distributions of the persons expected to consume the various wine purchases in the United States. The panel member and male head of the household were the dominant users of both varietal and nonvarietal table wines. Friends ranked third, and relatives fourth. In less than 3% of the cases, children were the intended users of varietal and nonvarietal table wines.



A summary of the percentage distribution of wine purchasers by age and type of wine is presented in Table 6. These percentage distributions are a direct reflection of the distribution of the households on the panel and the preference toward various wines by various age groups. All wine types except vermouth are purchased by all age groups. However, there is some notable skewedness to some of the distributions. A higher percentage of the older households purchased dessert wines, while a larger percentage of younger households purchased flavored wines. On a regional basis, there are some very notable exceptions which have significant marketing implications.




The percentage of households in each region making at least one wine purchase during the 12-month period was estimated to show the degree of regional market penetration (Table 7). Only 38.8%, or 4,469 of the 11,522 households on the panel, made at least one purchase of a wine product between February, 1975, and January, 1976. The 4,469 purchasing households made 21,219 purchases during the 12 months. The sale of wine in the United States is typified by low market penetration.



The regional market penetrations ranged from 52.6% in the Pacific region, to 19.6% in the East South Central region. Both the Pacific region and northeastern regions had a higher incidence of purchases than the other regions.

The fact that relatively few U.S. households buy most of the wine is shown by arraying the purchasing households according to total volume purchased. Figure 2 shows that 54.4% of the wine was purchased by only 10% of the purchasing households, or 3.8% of all the households on the panel. Two-thirds of the wine was bought by 20% of the purchasing households, or 7.6% of all the panel households. Wine sales are concentrated among a few households.



The affluent nature of wine purchasing households was revealed by the demographics of each decile. The average wife tended to be younger in the smaller purchasing deciles, while the family tended to be larger. In decile 1, the average male head of the household had at least a college education. In general, the years of education of the male head of household declined as the volume of wine bought decreased.

The most consistent demographic feature that segregates the largest and the smallest wine volume purchasing households was household income. The average household income was $16,724 in decile 1, the households buying the most wine. In the remaining deciles, the household income was less than $16,000 per year.

The average price paid per ounce was much less for the households buying large volumes of wine. The main reason for the lowest prices was the larger size of the wine container. In general, wine purchasing households, and especially the largest volume purchasing households, tended to be further along in their life cycles, as indicated by the ages of the wives, and had smaller families, more education, and higher incomes.


The largest wine company (Gallo) accounted for 32.9% of all wine purchases reported by the panel of households during the study period. The four largest wine companies (Gallo, United Vintners, Almaden, and Guild) accounted for 54.1% of all wine purchases. The 8 largest wine companies accounted for 64.5% of the reported wine purchases, while the largest 10 wine companies accounted for 66.7% of the wine purchases reported by the panel of households. The aggregated market share of the wine companies ranked second through tenth is only slightly greater (33.8%) than the relative market share of the largest wine company (32.9%).

The market shares reported above not only indicate the degree of market concentration in the U.S. wine industry, but also substantiate the validity of the household panel data in terms of market shares. The reported 1974 market shares published in Forbes (1974) are similar to those based on the reported household purchases, as indicated here:



To investigate brand preferences, the wines of the 10 largest companies in the United States were identified. The total quantity purchased of a given type of wine produced by each of the largest 10 companies was set equal to a base of 1.000. The total quantity of that given wine type purchased from other companies by the same households was expressed relative to the base of 1.000. In aggregating the data according to wine company, all the wine under all labels or brands of a given wine company was grouped for that company. If the sum of the index numbers for the volume of a given type of wine purchased from all other companies was less than 1.000, the panel of households was deemed as having a brand preference. Thus, a brand preference means that the panel bought more of the preferred brand than of all other brands combined.

The greatest degree of brand loyalties for varietal table wines was to United Vintners and Mogen David. This statement is based upon the fact that the households that bought a wine product by these firms did not buy a larger quantity of varietal table wine from all other companies combined than from the one under consideration.

Mogen David apparently had the strongest brand loyalty in the varietal table wine category. Households that purchased a Mogen David wine bought only 13.3% as much varietal table wine from other companies as they bought under Mogen David labels. This strong brand preference for Mogen David varietal table wines results partially from the fact that Concord wine dominates the varietal table wines produced by Mogen David, and the Concord grape has a unique taste and flavor.

The brand loyalty for nonvarietal table wines was weaker than that for varietal table wines. Only the households that bought Gallo and Canandaigua wine products bought less nonvarietal table wines from other companies than from Gallo and Canandaigua.

Panel households expressed brand loyalty or preference for dessert wines produced by Gallo, Guild, and Taylor. For sparkling wines, the panel of households revealed brand preference for Gallo, Franzia, and Guild. In the flavored wine category, only the flavored wines from Gallo and Mogen David had strong brand preferences. All of the Mogen David 20-20 wines were classified as flavored wine because some of them were not made from grapes and they varied in alcohol content from less than 14% up to 20%.

Paul Masson and Mogen David had the strongest brand preferences in the vermouth category. Brand preference in brandy purchasing was expressed for Christian Bros., Guild, and Canandaigua brandies.

In summary, there was some degree of brand preference for all types of wine. However, the panel of households did not show strong brand preference for all wines produced by a single company. Each wine company to some extent serves different segments of the wine market.


To ascertain how the market for wine in the United States was changing, the households completed a questionnaire during the last month of the survey. The questionnaire generated information relative to changing consumption levels, frequency of consumption, consumption patterns of other alcoholic beverages, and reasons and factors affecting the consumption and purchasing patterns.

About 15% of the households said they were drinking more wine than in the previous year. About 39% of the households reported drinking the same amount of wine, and 16% reported drinking less. Over 30% said they would never use wine. Of those that indicated that they never drink wine, the obvious question is why?

About 39% of those panel households that never drink said that it was against their religious or personal beliefs (Table 8). In the South, nearly half of the households never drinking wine maintained it was against their beliefs. The South, with the greatest percentage of nonbuyers, may be one of the most difficult areas in which to increase market penetration via marketing efforts. The Mountain region had the highest percentage of nonpurchasing households, 55%, who found wine consumption against their beliefs. In 45% of the nonconsuming households in the Pacific states, wine drinking is against their beliefs.



The western regions have the highest percentage of households that do not drink wine because they feel it is bad for their health. For all regions, about 13% of those households not drinking wine claimed it to be bad for their health.

Forty-six percent of the nondrinking households said they did not care for the taste of wine. Over one-half of the nondrinking households in the New England, Middle Atlantic, and East North Central regions listed taste as a reason for not consuming wine.

Eight percent of the nondrinking households said that wine was too expensive. Almost 20% in New England felt it was too expensive, while only 3% in the Pacific region listed wine as too expensive to drink.

About 38% of those who do not drink wine drink at least an occasional beer, and about 30% of the nondrinkers of wine drink some distilled spirits. Almost two-thirds of the households that never buy wine never consume alcohol.

Of the households drinking less wine, almost 75% said they now drink wine less often than once a month, and 10% consume wine no more than once a month. It would probably take a major marketing and merchandising program if these households were ever to become important in the wine market. These households are close to being nonpurchasers.


Households were asked to rank the five most significant factors in buying a wine. The importance to panel households of brand name in the selection of wine was paramount. Brand received more first place votes than any other factor listed. Almost all purchasing households listed brand as important, and a majority of those listed it either as first, second, or third choice. Perhaps brand loyalty is much stronger in some regions than that reported above at the national level.

Advice of friends and/or relatives was the second most important factor. Price was the third most important factor, and the label was fourth most important. Panelists often mentioned that the label provided information on using and serving wine. The fifth most important factor was flavor. Neither flavor nor menu were provided as choices on the questionnaire, but were written in the "other" category. About 10% of the respondents wrote in that flavor was the most important factor in selecting a wine. Advertising, advice of wine critics, and displays were less important.


Current purchasers are primarily the higher income and better educated households. The potential for larger wine sales appears to be significant if the industry can have a larger percentage of the households purchasing wine and with a higher degree of frequency. In addition, the wine industry may be overlooking the importance of sales to single member households and in establishments away from the home. Our data indicates that as much as 40% of the wine may be purchased by such individuals and/or in such establishments. Little marketing efforts have been targeted towards these market segments.


Folwell, Raymond J., and Baritelle, John L. (1978), The U.S. Wine Market, Washington, D.C.: U.S. Department of Agriculture, Economics, Statistics, and Cooperatives Service, Agricultural Economic Report No. 417.

Folwell, Raymond J., and Baritelle, John L. (1975), "Households Purchasing Table Wines in the United States," in Technical Assistance in Fostering the Economic Development of the Wine-Grape Industry, prepared for Economic Development Administration, U.S. Department of Commerce.