undefined

Introduction to conjoint analysis

Conjoint analysis is a category of research methods that mimic the respondent's real world tradeoffs when making decisions. Adaptive choice-based conjoint is one of the most advanced and tailored applications that learns from respondents as they answer questions. It is used for pricing studies, product optimization, healthcare options and many other things. To get a more in depth understanding of conjoint analysis, refer to our page on conjoint analysis.

Learn about conjoint analysis

What is adaptive choice-based conjoint (ACBC)?

Adaptive Choice-Based Conjoint (ACBC) is our most advanced system for conjoint analysis. ACBC is a newer approach to preference modeling that combines elements of CBC (Choice-Based Conjoint), artificial intelligence, and (optionally) dynamic list-building.

An Adaptive Choice interview is an interactive experience, customized to the preferences and opinions of each individual. It tends to probe more deeply into each respondent’s decision structure than a traditional Choice-Based Conjoint , but the survey is often twice to three times as long. Fortunately, respondents find the adaptive nature of the survey more engaging than CBC, so they usually perceive the questionnaire to be more enjoyable and to last about as long as the shorter CBC.

How does adaptive choice-based conjoint work?

Step 1 Build Your Own

Truck Conjoint Example1

ACBC's question flow incorporates the well-established theory that buyers make complex choices by forming a consideration set (typically using cut-off rules) and then choosing a product within that consideration set. We display relevant products for respondents to consider based on the preferred product that respondents first specify using a BYO (Build-Your-Own) exercise.

Step 2 Screening

Truck Conjoint Example2

In addition to the standard part-worth utilities that are useful for segmentation and market simulation, we capture the specific "must-have" and "unacceptable" rules that respondents express during the screening process. We also can tabulate the responses from the BYO (configurator) question. These data provide greater insight than typical CBC studies.

Step 3 Choice Tasks

Truck Conjoint Example3

ACBC is remarkably flexible. You can include the three major sections (BYO, Screening, Choice Tasks).

ACBC can leverage Lighthouse Studio’s dynamic list-building based on questions encountered earlier in the questionnaire. For example, you might be studying 24 total brands and only want to carry each respondent’s top few considered brands into the ACBC section of the survey.

When to use ACBC

  • Five or more attributes
  • When conjoint sections of about 7 to 15 minutes fit within the questionnaire

Why it works

Because of the BYO and screener sections, respondents find adaptive CBC interviews more engaging, realistic, and relevant compared to traditional (static) CBC interviews. Even though the interviews are typically longer than standard CBC questionnaires, respondents generally prefer the overall experience.

Want help with your first conjoint study?

Want in-depth help with your first project that goes beyond our free technical support? Our Sawtooth Analytics consulting team has deep expertise, decades of experience, and is ready to help.

Learn about our consulting team

Benefits of ACBC

  • More engaging, relevant interviews than traditional conjoint methods
  • Directly incorporates unacceptables and must-haves (non-compensatory decision-making)
  • Can probe deeper into each respondent’s choice hierarchy than traditional conjoint methods
  • Can work even with the smallest of sample sizes
  • Solid behavioral theory (consideration, then choice)
  • Solid statistical theory (near-orthogonal experiments and choice data)

Try ACBC in Lighthouse Studio

Lighthouse Studio

For developing web-based, CAPI (mobile devices not connected to the web), or paper-and-pencil (3rd party platform) CBC studies. This is installed, Windows-based software that seamlessly integrates with our web-based services for free survey hosting.

undefined