Due to the increasing competition in today’s consumer markets it is becoming more and more important for companies to meet customer expectations. In this study we report on a conjoint analysis that was conducted for the Virgil Company to determine which alternative is the best for them when entering into the wine category. As Virgil intends to enter into an existing market with a new wine, our aim is to provide the company with sufficient information about the consumer preferences to ensure a successful product launch. Insofar we will help Virgil to customize the wine so that it will have the highest value for the consumers.
The conjoint analysis will help us to estimate the possible impact of “Virgil’s wine” on the existing competitors’ market share. 2. Background The present study is conducting a conjoint analysis to determine which factors are the most important among customers when choosing wine. Our approach is to analyse the current market situation in the wine category in order to specify which wine attributes (region, price, and brand) are the most significant for the customers’ choice and thus which alternative is the best for Virgil to adopt. Conjoint Analysis is a multivariate technique to analyse customer preferences.
It is based on the hypothesis that each product consists of a bundle of capability characteristics (e. g. price, package, brand, and region). The better the test person evaluates the different attributes the higher is its preference for this product and thus the higher is the probability that the product will be bought (Schaupp ; Belanger 2005). Conjoint analysis is used to evaluate consumer preferences; thus its objective is to investigate an optimal ‘product package’ that consists of different attributes levels (popular region or less popular region with well known brand or unknown brand for example).
Then the result of the study will provide us with the perfect combination of attributes and their levels. The conjoint analysis has not sufficient capacity to predict a precise impact of new alternatives within a product category. This means that the number of attributes used in a conjoint analysis will never reflect the actual amount of attributes that affect the consumer in the choice process (Gibson 2001, p. 18). It is also impossible to add all possible attributes and levels to a conjoint analysis because such an amount of alternatives would make the result unreliable as the “respondent fatigue” (Wyner 1992, p.47).
Prior to the analysis researchers have to choose the most important attributes for the analysis itself in order to keep it simple and assessable (Gibson 2001, p. 18). This is a decisive problem as the result of the conjoint analysis normally should provide information about the importance of attributes. The whole process would be useless if researchers would know the most important attributes before they had conducted a conjoint analysis. This shows that the result highly depends on the researcher or on the person who is selecting the attributes for the analysis so that it can’t produce precise estimations.
Taking into consideration that the whole process is first of all based on the ability of researchers to predict consumer choice it is clear that different research groups could come up with different results. Furthermore researchers must decide prior to the analysis which attributes have the highest value for those who are “at the cusp of choice” (Gibson 2001, p. 18). This situation evokes two main questions: i. What is important for consumer at the cusp of choice? ii. Which consumers are at the cusp of choice? These two points are a further example of the complexity of the estimation before a conjoint analysis.
Therefore it is clear that it is impossible to make a completely precise estimation prior to the study. Moreover, conjoint assumes that the attribute or the level with the highest importance for the consumer is decisive for the brand choice. This would mean that if conjoint found out for instance that it is of high importance for the customer that mobile phones are small the consumer will choose a brand regarding this attribute level. But if all mobile phone manufacturers produce small phones the result of conjoint would lose value as mobile size wouldn’t affect the brand choice (Gibson 2001, p. 18).
Finally conjoint is not considering that consumer’s choice is also affected by personal and individual perception of the product attributes. Some people might find that a Coke is refreshing but others not for example. It is also clear that people are not always rational and buy the good with the highest personal value for them (Wyner 1992, p. 46). A conjoint analysis is losing significance because of the fact that it is based on estimations prior to the study and on some wrong assumptions about the brand choice of consumers so that it is not a precise model to estimate the impact of new alternatives within a product category.
This study conducted a conjoint analysis to find out which alternative is the best for Virgil to adopt when entering into the wine category. We found that a change regarding the brand attribute has the highest impact on the respondents. Therefore it is also reasonable that the alternative with a “well known brand” combined with a lower price and also well known region has the highest potential market share and should be adopted by Virgil. 7. Future Research It is clear that this study is based on a simple model that does not include the financial aspect of production.
Virgil success in the wine category depends apart from “the right attributes” of the product on their ability to produce a high quality wine (well known brand & region) to a lower price. Another point which also has to be considered in further research is the behavior of the customer. A study needs to examine how many customers are willing to switch to Virgil’ wine after it is available in the market.
This is very important because the conjoint model assumes that the consumer will always buy the good with the highest individual value. In reality we will have to consider the loyalty of customers. It is also important to know how many people at the cusp of choice would rather buy Virgil’s wine than other.
Gibson, LD 2001, ‘What’s wrong with conjoint analysis’, Marketing Research, vol. 13, no. 4, pp. 16-19. Wyner, GA 1992, ‘Uses and limitations of conjoint analysis – part II’, Marketing Research, vol. 4, no. 3, pp. 46-47. Lonial, S, Menezes, D ; Zaim, S 2000,’Identifying purchase driving attributes and market segments for pcs using conjoint and cluster analysis’, Journal of Economic and Social Research, vol. 2, no. 2, pp. 19-37.