Conjoint Analysis – An Elucidation



Suppose you own a company and your company is about to launch a new product in the market or, maybe you are about to re-launch an existing product with certain modifications.

Then, you can go on with a bold front having full faith on your business strategy or can be wise and do a little market research. Otherwise, you might end up with a product which no one wants to buy. Success of a product depends upon a lot of variables. First and foremost, your product needs to be perfectly functional. A lot depends on your promotion skills, how you reach out to your customers with your product and how good you are at convincing them to buy your product. Then, there are other important aspects such as packaging, material, etc. Even information about the place of manufacture of the product can prove to be vital. For instance, when you are buying a laptop what features do you take into count? A good configuration would be on the top of your priority list. But, you would check other aspects too, such as battery backup, sleekness, buyer’s review etc. Any of these features might be the primary attribute for you but you would also take the other aspects into count.

When so many factors come in to play, choosing the right combination entirely based upon guesswork would be a bit too much of a risk. So, how exactly should you plan your business campaign keeping in mind all these attribute? That is exactly what ‘Conjoint Analysis’ does for you.

What is Conjoint Analysis?

Technically Conjoint Analysis is a statistical technique which is most commonly used for doing market research, product management and operations research in order to determine exactly what is the customer looking for when he or she avails an individual product or service. The process is a special case of regression analysis and it does not have any clear statistical definition. Some of the following points can be applicable to conjoint analysis.

  • The task of data collection lies upon multiple individuals, while each individual has multiple data points of their own. This makes the process a multi-layered system.
  • A choice of tradeoff situation is reflected by the dependant variables.
  • The independent variables are definite thereby, hinted as binary numbers (0,1).

How does the Conjoint Analysis Work?

First of all, the research participants are required to make a series of tradeoffs which upon being analyzed, reveals the comparative significance of component features.  Research participants should be set into same segments on the basis of values, objectives or any other factors in order to improve the predictive analysis of the tradeoffs made by them. This survey can be administered to evaluate the demands of its respondents in various ways sometimes, as a ranking exercise and at times, as a rating exercise. In this process, the respondents are supposed to award each trade off scenario with points according to them. The tradeoffs are often represented as consumer choices where a consumer usually chooses a preferred alternative from a selection of competing alternatives. It can also be a sum allocation exercise and mostly used in pharmaceutical industries.

With the increase in number of attributes, the Conjoint Analysis procedure might face information overload affecting the validity of the experiment under such circumstances. This kind of outcomes can be avoided with the use of Hierarchical Information Integration. For instance, you have a plot of land and you are planning to build a housing complex. The plot happens to be situated close to a University. A market research firm is hired who conducts a survey to weigh the market demand. The students of the University are given a couple of cards containing the various attributes of the housing complex. The students are then asked to arrange the attributes from least important to most important according to them. This exercise brings out the priorities of individual students and helps build the weighted score of the attributes.

Types of Conjoint Analysis

  • Previously, Conjoint Analysis involved full profile studies. A batch of attributes is included within these studies and the respondents are given handful of such profiles to evaluate these attributes. The implicit utilities for the levels can be calculated using dummy variation regression analysis. One drawback of this method is that the number of allowable attributes is highly restricted. If the properties are too high, the task on the part of the respondents becomes too big. Even with fractional factorial designs, the number of profile evaluation can increase rapidly.
  • In order to avoid the above mentioned problem and increase the number of attributes to above 30, Hybrid Conjoint Analysis methods are used. Common alternate techniques include some form of self-explication before the conjoint tasks and some form of adaptive computer-aided choice over the profiles to be shown.

Article signup banner

Process of Information Collection

The data used for Conjoint Analysis is generally gathered through market research survey. This data could also be collected from a test market experiment and then, Conjoint Analysis is applied to the configuration. While designing Conjoint Analysis interviews market research rules of thumb is applied with regard to statistical sample size and accuracy.

The number of fields in the questionnaire depends upon the number of attributes which need to be considered and also the type of Conjoint Analysis used. A typical Adaptive Conjoint questionnaire takes something around 20-25 and at times, even 30 minutes to be completed. Whereas, in case of a choice based conjoint, a small profile set is used and distributed across the sample. This takes around 15 minutes to be filled. Choice exercises may be displayed as a store front type layout or in some other simulated shopping environment.

Positives of Conjoint Analysis

  • Individual preferences are efficiently measured.
  • It uncovers real or hidden drivers which may not be apparent to the respondent themselves.
  • Realistic choice or shopping task
  • Able to use physical objects


Conjoint Analysis is not devoid of drawbacks. For example, designing conjoint studies can be complex and with too many options, respondents resort to simplification strategies. It is difficult to use for product positioning research because there is no procedure for converting perceptions about actual features to perceptions about a reduced set of underlying features.

Other tools you might find interesting