Each cluster was compared on a range of demographic (age, gender, income) and wine consumption behavior and attitudes, but no major differences were found between the groups — an interesting and very important result.

Typical marketing and wine marketing professionals have been using demographics and other easy-tomeasure variables to segment potential consumers for their products.

We have now been using “choice” experiments for more than six years, and continue to find that demographics and attitudes do not predict how consumers choose wine. Consumers definitely use different strategies to make wine choices, and our experimental results have strong correlations with actual sales in the market, so we believe wine marketers need to reconsider how they segment consumers.

As a result of this finding, producers cannot specifically target the five consumer segments by supplying different products to different sales channels. However, this study outlines certain combinations of wine attributes that are valued by certain consumer groups.

For example, lower-priced wines should display information on food pairing and elaborate taste descriptions, which are of high utility for about one-third of consumers (C1). A smaller share of about one-fifth of consumers who value environmental back label attributes like to read history and production method information and elaborate taste descriptions (C2).

Wines sold at medium and higher price points are recommended to display food pairing, elaborate taste descriptions, and winery history information on back labels (C3 and C4). All of these additions had positive impacts on choice probabilities.

The low importance of an environmental message agrees with findings by H. Remaud et al., who found that only 15% of Australian wine consumers consider environmental claims when making a wine purchase decision.4 In this study, it only has a notable impact for one consumer cluster (C2).

While almost all information on back labels had a positive impact on consumer choice, ingredient information had a strong negative effect for about one-third of frequent Australian wine consumers.

For a small segment (13%), the negative impact on choice of 59% implies that a very positive attribute such as a very low price would have to compensate for the substantive negative reaction to ingredient information. This strong aversion to the list of ingredients was surprising, and cannot be explained completely by this study. It is possible that the mention of complex and unfamiliar ingredients creates a stronger feeling of risk or perhaps is inconsistent with the overall image of wine as natural and healthy.

These findings have important implications for a government considering compulsory labeling of wine ingredients as recently supported by Australia’s leading consumer organization, “Choice.”3 The government has to ensure that consumers are educated and informed about the meaning, risk, and potential health impact of those ingredients and needs to find terms and language understandable to buyers. Otherwise, these information measures are likely to have an adverse impact by creating risk perceptions instead of reducing consumer uncertainty.

wine choice decision, so these findings must be combined with a good understanding of front labels to be effective. We did find that, except for ingredient labeling, these back label statements had either a positive or no effect, so using these results to fine-tune a back label is recommended.

We were unable to measure the interactions of different back label statements over the total consumer sample. Some combinations of statements had a stronger effect for some clusters than for others. Back label statements are an inexpensive and efficient means for small- and medium-size wineries to interact with consumers, and more attention should be paid to what is actually printed on the back label.