Home >> Software Help >> Trade-Off Analysis / Conjoint Analysis
View Demo | Download Tutorial | Frequently Asked Questions | Glossary of Terms

The Model

Conjoint analysis derives from the field of psychometrics, or the quantitative analysis of human judgement. Since its introduction to marketing research, its use has grown rapidly and has become one of the most popular procedures for measuring customer preferences across a variety of products and industries. The goal of conjoint analysis is to use a person's evaluations of alternative products to discover the structure of their underlying needs and customer value. From this structure, estimates can be developed about how the buyer might respond to different products designed for these needs.

The procedure works as follows:

  • Product concepts are designed using features (e.g., price) with various levels (e.g., $1, $2, $3) that are believed to influence the buying decision.


  • The various combinations of the levels of the features for each product concept (e.g., green color, $2, low quality) are selected from an experimental design (or plan).


  • A potential buyer is asked to provide an overall evaluation about each alternative product. This evaluation can be a rated preference, intention-to-buy, or another predisposition measure. It can also be a preference rank ordering of all the product concepts evaluated.


  • The evaluations are then mathematically decomposed into utility scales (based on product features and their levels), which can then be recombined to produce the original evaluations.


  • The utility values, which reflect a buyer's underlying needs, can then be used to determine the relative importance of a product feature as well as buyer sensitivity to different levels of the feature.


More detailed explanations on the mathematical model behind the Conjoint Analysis can be found in Chapter 2 : "Customer Value" of the Harvesting Customer Value Book.

To see a list of features included in our Conjoint Analyis software, Click here.


Getting started

Before you start running our software, please take a look at the Compatibility List.
Due to our intensive use of various Web technologies you may experience difficulties running a software in your browser.
We apologize for any inconvenience this may cause and are working on full browser compatibility.

You may have to enable Javascript, Cookies and/or accept Active X Controls. Please refer to the Compatibility List.

Before running the Conjoint Analysis software, you may need to configure your browser to accept ActiveX software.
Instructions can be found here.

Read this help section and the Tutorial carefully before running the software.

You can download here (kirin.zip) an example file (Kirin beer Case) you can load into the software.


Running the software

Step by Step instructions on how to run the software can be found in the Tutorial.


Understanding the results

Below a brief summary of the Steps in Designing and Executing a Conjoint Study:



Stage 4 involves using the individually obtained part-worth utilities as the basis for segmentation. This stage usually involves clustering these data, followed by cluster-profiling. Segments based on conjoint data can provide great insight as they are based on very careful assessment of customer value defined over critical product attributes and levels.

Conjoint analysis is one of the most widely used modeling techniques in marketing, with over 25 years of use. It involves designing a product option from attributes and levels and obtaining customer evaluations of these options. These evaluations can then be decomposed into basic attributes and levels that are meaningful to customers and that managers can use to design new products and services. In particular, we can segment customers into groups whose utilities (values) within groups are similar and which are different between groups - an important criterion for meaningful market segmentation.

Yet conjoint analysis must be used with care. It is a compensatory model - that is, high values on one attribute compensate for low values on another. However, many consumer choice situations are non-compensatory. For example, customers may require that a digital camera must have an auto-focus capability before they would ever consider buying it. If this is the case, the conjoint model violates the compensatory assumption about customer behavior and the results will not be predictive of actual market choices. In addition it is vital to include the appropriate set and levels of attributes. If critical dimensions are missing or in error, the results will be suspect.