What is Conjoint Analysis?

Conjoint (trade-off) analysis is one of the most widely-used quantitative methods in Marketing Research. It is used to measure preferences for product features, to learn how changes to price affect demand for products or service, and to forecast the likely acceptance of a product if brought to market.

Rather than directly ask survey respondents what they prefer in a product, or what attributes they find most important, conjoint analysis employs the more realistic context of asking respondents to evaluate potential product profiles. Each profile includes multiple conjoined product features (hence, conjoint analysis), such as:

Tourism

Tourism Example

Source: Orme, Bryan (2015), “Perceptual Choice Experiments: Enhancing CBC to Get from Which to Why,” Sawtooth Software Conference Proceedings, Orem, UT.

Health Economics

Health Economics Example

Source: Saigal, Christopher and Ely Dahan (2012), “The Voice of the Patient,” Sawtooth Software Conference Proceedings, Orem, UT.

Agricultural Economics

Agricultural Economics Example

Adapted from: Banerjee, Swagata “Ban”, Steven W. Martin, and Darren Hudsen (2006), “A Choice-Based Conjoint Experiment with Genetically Engineered Cotton in the Mississippi Delta,” Southern Agricultural Economics Association Annual Meetings, Orlando, Florida, February 5-8, 2006.

There are different ways to show product profiles. The most popular technique used today is Choice-Based Conjoint, or CBC, as seen in the examples above, where respondents are shown multiple product concepts at a time and asked which option they would choose. Sawtooth Software’s CBC module may be used to create exercises like this. Prior to Choice-Based Conjoint, the original version of conjoint analysis was developed in the early 1970s, which shows products one-at-a-time. Sawtooth Software's CVA module may be used for this traditional form of conjoint analysis. Later forms of conjoint analysis showed products in pairs (CVA or ACA for Adaptive Conjoint Analysis), or sets at a time (for CBC or ACBC for Adaptive CBC).

In a conjoint exercise, respondents usually complete between 12 to 30 conjoint questions. The questions are designed carefully, using experimental design principles of independence and balance of the features. By independently varying the features that are shown to the respondents and observing the responses to the product profiles, the analyst can statistically deduce what product features are most desired and which attributes have the most impact on choice. In contrast to simpler survey research methods that directly ask respondents what they prefer or the importance of each attribute, these preferences are derived from these relatively realistic tradeoff situations.

The result is usually a full set of preference scores (often called part-worth utilities) for each attribute level included in the study.

Conjoint market simulators let the researcher define specific competitive contexts (specific products in competition with another) and project the share of choices (shares of preference), given respondent’s estimated part-worth scores. These simulators let researchers and managers test a variety of what-if scenarios.

Market simulators can be taken one step further. Rather than using them to answer the question: "How good is this product?" they can be used to discover "Which product is best?". Computer search routines (such as Sawtooth Software's Advanced Simulation Module) can efficiently find optimal products, based on the criterion of utility, share, revenue or profit.

Thousands of conjoint analysis studies are conducted each year—over the internet, on mobile devices, and through person-to-person interviews. Organizations use conjoint analysis to make better pricing decisions, design new products or line extensions, reposition products, and save on research and development costs.

With 30 years in business and thousands of users worldwide, Sawtooth Software is the consensus leading provider of conjoint analysis software. Companies like Procter & Gamble, Google, General Motors, General Electric, and Microsoft use our software and have presented case studies and success stories at our popular research conferences.

Sometimes it can be challenging to decide which conjoint method is most appropriate for your particular research situation. We've developed an interactive advisor to help you decide which conjoint method might be best for your specific situation. We’ve also written a document entitled Which Conjoint Method Should I Use? for further reference.

Want to talk conjoint?
801-477-4700