Choice-Based Conjoint Analysis Guide [Example Questions and Case Study]

choice based conjoint analysis

In this article, we take a look at the benefits of choice-based conjoint (CBC), how and when to conduct a CBC study, what CBC questions look like, and an example of a CBC project.



Table of Contents: 




What is choice-based conjoint analysis? 

Choice-based conjoint analysis (CBC), also known as discrete choice modeling, is an advanced market research method that identifies consumers’ preferences when considering a product or service. This is done by asking research respondents to make trade-offs between competing products, each of which has a variety of attributes. Asking consumers to choose their preferred product reveals the importance of different attributes in determining consumers’ willingness to pay. Product attributes might include brand, design features, price, or style; attribute levels (within each attribute) might be Ford and Toyota, built-in nav system, heated seats, sporty or family style, etc.

CBC is the most commonly used type of conjoint analysis. It differs from other conjoint approaches in that it presents consumers with full product profiles (rather than just asking them to rate attributes separately, as in two-attribute trade-off analysis) and it allows for the inclusion of price as a determining attribute (which is not an ideal use case for another type of conjoint - adaptive conjoint analysis; this type of conjoint changes as each person answers the survey questions to consider their individual preferences).

Back to Table of Contents




Benefits of a choice-based conjoint study 



Because respondents are presented with profiles that detail the different attributes contained within each product, this method mimics a real-world purchase scenario. Buying a product can be a complex process, with subconscious decisions made along the way, so asking respondents to make product trade-offs reveals which attributes and attribute levels truly drive the purchase decision.

Attribute valuation

Traditional surveys ask respondents to rank or rate attributes and attribute levels, which can give an indication of how important different features are when consumers make purchases. However, the problem with considering attributes in isolation is that this lacks the contextual information required to assess how likely a purchase will be. For example, bread buyers might say that they rate whole grains, added vitamins, and bread softness highly, but it can be difficult for them to say which of those attributes is more important than others.

Forcing consumers to make trade-offs reveals the relative importance and value of each attribute. Some product profiles in a CBC might not even be chosen at all, revealing attributes that are of little or no importance to consumers (and therefore, not worth the investment). Each attribute’s value metric is known as a part-worth utility score, and is calculated for each attribute level in a study. This is a great springboard for a needs-based segmentation that defines what different consumer groups are looking for in a product.

Knowing each attribute’s valuation is helpful for designing products and ensuring the whole product package is attractive to consumers. After determining the top attributes, you can use a conjoint analysis to ensure that when these different parts are combined, the product is still overall appealing.

The effect of price

Determining the optimal price level for a product is incredibly important, but can be difficult to measure in a research study as respondents will almost always say that price is important to them and that they want the lowest price possible. Further, price isn’t a fixed attribute with a limited number of attribute levels - it can always be increased or (to a degree) decreased.

With choice-based conjoint, brands can test out different price levels in their product profiles to identify the overall price range that consumers will consider buying their product. This is one of the best ways to identify a ‘fair’ and justified price to charge for your products. Beyond determining price level, conjoint can also project revenue modeling to find a sweet spot between a price index and actual revenue.

Respondent enjoyment

Because a conjoint survey feels like a real-life scenario, respondents enjoy the ability to choose between different product profiles rather than simply answering questions about separate attributes or giving individual rankings/scores.
Back to Table of Contents



How to execute a choice-based conjoint analysis 

There are a number of factors to consider when designing and conducting a CBC survey. At the design stage, brands need to decide on the following:

  • Sample size: This needs to be big enough to provide meaningful data on consumer preferences. The number of respondents needed will depend on the complexity of the design, but a general guideline is to have a few hundred respondents for each product profile you’re measuring.

  • Choice type: Choose how respondents will evaluate each set of product profiles (i.e. combinations of attribute levels). You might want to force respondents to make a single choice from the sets of products shown or provide them with a ‘none of the above’ option.

  • Number of profiles per set: Decide how many product profiles should be shown per set. Too many profiles can become tedious for respondents, while too few profiles won’t provide enough comparative data.

  • Number of sets per respondent: Similar to the above, decide how many overall ‘sets’ of products each respondent will evaluate so that respondents aren’t overwhelmed but still provide enough data for analysis.

  • Attributes: These are the features of each product or service you’re researching. These might include price, size, color, brand, and style. Aim for no more than six attributes to avoid overloading respondents.

  • Attribute levels: The variations within each attribute - such as large, medium, and small; blue, red, white, and yellow. Again, aim for no more than six levels to keep respondents engaged.

Once the above factors have been established, a brand will launch their choice-based conjoint survey amongst their target audience. Randomized choice sets of product profiles are shown to each respondent, and respondents choose their favorite product from each set. At the analysis stage, each attribute level’s part-worth utility is calculated, as well as its relative importance to other attribute levels in the study.

quantilope offers a fully automated approach to CBC, from survey design to final analysis. Its survey templates and pre-programmed CBC method ensure all relevant information is included in a conjoint study. quantilope’s CBC analysis output includes a market simulator that projects how different product profiles would be received by consumers in the real world and identifies propositions with the highest consumer appeal and willingness to pay.
Back to Table of Contents



When to use choice-based conjoint analysis for your business   

Choice-based conjoint analysis is used across a broad range of business areas, from consumer packaged goods (CPG) to services and healthcare. Wherever there is the possibility of different product or service propositions, CBC is an excellent way to determine which profile would be most appealing and profitable.

If your business wants to explore any of the following, CBC is a great methodology to leverage:

Projected market share

If you have an idea for a new product or a revamp of an existing one, it pays to know whether it will sell well once launched. A common use of CBC is to determine which feature combination will claim the largest market share.

Nailing down product features

If you’re at the product development stage and have an idea of features for your product but don’t know which will be most important to consumers, CBC will tell you which ones, and with which combinations, to include for maximum consumer appeal.

The right price for a product

A crucial question for any business is how to price its offer. With CBC, each product profile can include price as one of the attributes and the analysis will reveal the perceived value of product benefits (i.e. what consumers are willing to pay for the features a product has). It will also give a good idea of price sensitivity - i.e. how a product’s demand is affected by price - and how market share will affect revenue at different price levels.
Back to Table of Contents



Examples of choice-based conjoint analysis questions  

A choice-based conjoint analysis questionnaire can look different depending on the product or service being tested, or on the survey platform used, but the general principle is always the same.

If you’re conducting a survey on smartphones, your choice sets presented to each respondent could look something like this:

Brand Brand A  Brand B  Brand C
Screen size 6.7 inches 8.2 inches  6.1 inches
Camera Ultra-wide  Dual-lens Autofocus
Storage 254 GB 128 GB 512 GB
Battery life 29 hours  20 hours  23 hours



The smartphone profile that a respondent opts for will give an insight into which features they place the most importance on. For example, they might sacrifice a better camera for a longer battery life, or choose a larger screen despite lower storage.


As another example, a conjoint question for hand soap could include the following attributes:

Brand Brand A  Brand B  Brand C None of these options
Size 750ml 500ml 300ml
Scent Orange blossum Eucalyptus Honey
Antibacterial No Yes  Yes
Price $9 $16 $12



Will respondents go for an antibacterial hand wash, whatever the price? Is size a key factor because they have a large family? Are all these profiles too expensive for hand soap, and a respondent would choose 'none of these options'?


As a third example, a restaurant conducting a conjoint questionnaire might include the following attributes to see which menu items are most appealing:


Main type of cuisine Pizza Steak  Vegetarian
Cocktails Yes Yes No
Family-friendly Yes No Yes
In the town center No Yes  Yes
Average meal price $20 $45 $32


If the restaurant were planning a new menu, the conjoint data would help narrow in on which menu items are most appealing, what the atmosphere should be like, and how they should price their meals.
Back to Table of Contents



Example of choice-based conjoint analysis study  

quantilope’s automated choice-based conjoint analysis is a popular methodology for many platform users and clients. One such client is PAX, a leading global cannabis brand that wanted to gather consumer insights around a new product offer in a growing market. Product design and innovation were essential to PAX’s growth, so using a CBC to explore product formats and benefits was key.

Using quantilope’s CBC, PAX was able to present a range of product possibilities to consumers and, by means of automated analysis, understand attribute importance and benefit configurations that would appeal to the most consumers.

“Two weeks after we signed on with quantilope I got a direct request from our CEO to run a Conjoint analysis. I would not have been able to do it without quantilope; my other option would have been to find a specialist and lose time requesting and reviewing proposals.”
-Kristen Archibald, Sr. Consuemr Insights Manager at PAX


For more on this successful conjoint study, access the full case study here.
Back to Table of Contents




How quantilope can help with your next choice-based conjoint analysis 

quantilope’s expertise in AI-driven methodologies (including choice-based conjoint analysis) provides brands with the confidence needed to design a successful product or service offer.

Although CBC is a sophisticated and complex research method, quantilope makes the process seamless and straightforward. Simply select the CBC from quantilope’s list of pre-programmed advanced methodologies, design your product profiles with various attributes and attribute levels, ‘configure’ the remaining setup in one click, and set your survey live.

Review your conjoint analysis data through a variety of charts that show things like an optimal price point, acceptable price range, average part-worths, individual attribute importance, and more. Then, merge all your findings into one interactive, shareable dashboard with automated significance testing.

For more information on how quantilope can help your business test new product profiles and features through automated choice-based conjoint analysis, get in touch with us below!

Get in touch to learn more about choice-based conjoint!



Related Posts

What Are Brand Perceptions and How Can You Measure Them?
Read More
How Can Brands Build, Measure, and Manage Brand Equity?
Read More
How To Use a Brand Insights Tool To Improve Your Branding Strategy
Read More
quantilope's 5th Consecutive Year as a 'Fastest Growing Tech Company'
Read More