Creating a CBC

Choice-Based Conjoint (CBC) is used to study how people make choices. It is the premier tool for optimizing features and pricing for products/services. CBC exercises show respondents realistic-looking scenarios where they are asked to make choices among different product concepts. Each of these concepts are made up of two or more attributes (or features). In making choices, your respondents make tradeoffs similar to those they make in the real world, revealing what drives their decisions.

By observing respondents' choices to such scenarios, we can build models that capture the preference weights (utilities) associated with those attribute levels that are driving those choices. With that information, you can simulate (using the market simulator) what people within a given market would choose when facing a multitude of potential choice scenarios.

Click any Add button.
Select Choice-Based Conjoint.
Enter Question text.
Enter Attributes. Enter these one by one or paste in a list from a program such as Microsoft Word or Excel. When a list is pasted, a section for each attribute is created along with empty levels. There must be at least two attributes.
Enter Levels. Enter these for each attribute one by one or paste in a list from a program such as Microsoft Word.

Steps 3 through 5 are shown in the software.

In the real world, buyers usually aren’t required to buy a product or service—they can walk away and buy when there are better options/prices presented to them. The none option in CBC surveys allows you to mimic that real-world experience, by providing an opt-out choice. There are two types of none options: traditional and dual-response. By default, the traditional none is used in the CBC exercise.

Select the none type in the question settings.

The traditional none is enabled by default; it is shown as an additional option presented in each task. It includes the text "None, I wouldn't choose any of these." Respondents have the choice to select one of the product concepts or opt for none of them. This mirrors real-world scenarios where buyers are not obligated to make a purchase if they do not like any of the available options. The traditional none option can also be used for specifying a fixed alternative such as a status quo choice or a fixed alternative such as “I would prescribe diet and exercise to the patient.”

A traditional none option is shown below a CBC task asking respondents to select a vacation package.

In contrast to the traditional none, the dual-response none is an additional question presented after each forced choice task (where respondents are required to choose the best among multiple product alternatives). By default, it includes the text "Given what you know about current offerings and your budget, would you actually purchase the option you chose above?" and provides two response options: yes or no.

A CBC with a dual-response none asking respondents

Substitute the word “option” in the default message to "product," "package," or whatever term best reflects what the respondent is choosing in your specific study. This customization will add additional clarification to the respondent in their decision-making.

Though “Yes” and “No” are the default options, the verbiage or number of items can be changed. Another suggested way to ask the question is “How likely are you to purchase the product you selected?” and to provide a five-point scale such as:

Definitely would
Probably would
Might or might not
Probably would not
Definitely would not

Many researchers have experience with how the five-point scale relates to actual purchase/adoption in their product categories. Thus, the five-point purchase intent scale may be a familiar option for the dual-response none. That said, using scales such as the five-point scale requires the researcher to specify which response options indicate a “buy” or “no buy”, such as the top two boxes indicate a “buy” and the bottom three boxes indicate a “no buy” (described further below).

In most marketing-related CBC applications, the dual-response none substantially increases the likelihood of respondents choosing "None," which many researchers argue better reflects actual purchase intentions compared to the traditional none option which can understate the “None” rate. Importantly, no information is lost when respondents select "None" because they are first asked to choose the best product concept in the choice task. Therefore, the dual-response none acts as a safety net. It allows us to estimate a None parameter without concerns about potential loss in precision of utilities for the attributes and levels of interest, especially if the incidence of "None" usage is high.

Dual-response none exercises will export additional tabs intended for our desktop products in the design and choices download. Learn about the design & choices export.

This threshold is used to tell utility estimation which response options are considered as “yes, would buy” and which are “no, would not buy.” If desired, this can be changed after collecting data; however, it will require a republish to the survey.

When there are two items in the list, select which one is the "would buy":

The UI showing the numeric input to enter which item will be considered

When there are multiple items in the list, define the range of items to be considered as "would buy":

The UI showing two numeric inputs to define the range.

This counter shows the respondent the progress they have made through the exercise. This is different from the progress bar shown below the question which shows the progress through the whole survey. The counter can be turned off and the label field is not required.

The task counter is right below the question text.

This setting controls how the attribute labels appear to respondents. By default, this is set to Display to the side of concepts, however there are two additional options: Display inside concepts, and Not shown.

Display to the side of concepts

Display inside concepts

Not shown

If Discover detects that the respondent's screen size is below 834 pixels (smaller than an iPad), then it will automatically change the layout of the CBC exercise. Attributes will appear inside the concepts and each concept will take up the width of the page and scrolled through in a carousel. This is done to ensure the respondent can easily see all the information for each concept. If you would like to change the behavior of this, there are options to Swipe concepts vertically or Maintain desktop layout. When concepts are shown horizontally or vertically, dots are shown to indicate which concept is being viewed. When a concept is selected, the dot changes to look like a radio selection.

Swipe concepts horizontally

Swipe concepts vertically

Concepts are swiped through vertically. Dots are shown on the left to indicate which concept is in frame.

Maintain desktop layout

The select button has three states: default, hover, and selected. Default and hover states appear to respondents when they are looking at a CBC task or hovering their mouse over a concept. The text shown is the same in both states, however the color of the button changes from white to blue, so be cautious if you choose to change the text color. The selected state appears once they click a concept.

Discover Help Guide

Creating a CBC

Introduction

Creating a CBC exercise

"None" option

Traditional

Dual-response

The “Would buy” threshold

Exercise formatting

Show exercise progress (number of tasks) & task label

Attribute labels

Mobile layout

Select button

Related Content