Conjoint analysis is a method used primarily to study consumer preferences.
Conjoint analysis
Optimization and development of the product offers,
market modelling
It belongs to quantitative research methods and is used in the development of new products or in the adjustment/modernization of existing ones, when it is necessary to determine technical, consumer and other characteristics a product requires to stay in high demand.
Joint analysis is one of the most complex marketing research methods. It is distinguished by multi-stage structure, complexity of tools, sophisticated methods of statistical analysis (e.g. regression analysis). The purpose of the method is to identify product characteristics most important for consumers, in order to obtain an “ideal product” profile.
The advantage of the conjoint analysis is simulation of the choice of goods made very real when responding to the questionnaire. This is achieved due to the fact that in the questionnaire the respondent is offered to choose from alternatives containing a comprehensive description of the products, each alternative containing a different set of product characteristics (“product profiles”).
For example, in a study of tea products, “product profiles” to choose from may look like this:
| Leaf | Package | Price | Weight | Country | |
|---|---|---|---|---|---|
| Profile 1 | pekoe | box (cardboard) | 350 y.e. | 200 g | Ceylon |
| Profile 2 | large leaf | box (metal) | 400 y.e. | 200 g | China |
| Profile 3 | granulated | box (cardboard) | 300 y.e. | 200 g | India |
| Profile 4 | tips | teabags | 350 y.e. | 400 g | Pakistan |
| Profile N | … | … | … | … | … |
The names of the columns in the table (leaf, package, price, weight, country) are the so-called “attributes” (product parameters), and the contents of the columns are so-called “attribute levels” (specific values of the attributes). For example, for the attribute “country” the levels are China, Ceylon, Pakistan and India.
Profiles are formed as a combination of levels for each of the attributes, so the number of profiles can reach several dozen. Depending on the analysis scheme, the respondent evaluates the attractiveness of each profile according to the proposed scale (for example, from 1 to 100), or chooses between several proposed profiles (between two, three or four), or ranks the profiles according to the degree of preference (from the most attractive to the least attractive).
At the output, we get numerical coefficients showing how important each of the “levels” (the so-called “utility levels”) and each of the “attributes” (the so-called “importance of the attributes”) are for consumers. Knowing these indicators, we can quantify various combinations of product characteristics: the higher the total coefficient of a product with this combination of characteristics, the more attractive this product is for the consumer.
