What Is Conjoint Analysis? How It Works and When To Use It
Conjoint Analysis is a sophisticated market research method that guides businesses on which product or service profile will be most successful for them.
Table of Contents:
- How does conjoint analysis work?
- When to use conjoint analysis
- Different types of conjoint analysis
- How to create a conjoint analysis survey, with example questions
- Alternatives to conjoint analysis
How does conjoint analysis work?
Conjoint analysis is a technique where respondents are presented with a set of product or service concepts and asked to choose their preferred one. Within each description are multiple features (attributes) of that product/service, and options that can be compared on a like-for-like basis.
For example, if you were researching toothpaste you might present some of the following options for 100ml tubes with different price points, flavors, and benefit claims:
- Colgate - $3.70 - spearmint - plaque removal
- Crest - $3.25 - fresh mint - whitening
- Sensodyne - $4.20 - cool mint - gum health
Respondents are asked to make trade-offs between multiple products. However, the number of attributes included within each product description should be limited - ideally no more than six or seven - so that decision-making isn’t too complicated for the respondent; especially because in a real-life shopping scenario, consumers generally would not compare any more than this number of attributes when making a purchase decision.
When to use conjoint analysis
Conjoint analysis is invaluable for any research into the impact of different product features on consumers’ purchase intent. If asked directly, in either quantitative or qualitative research, consumers will often say that all attributes are equally important, or won’t be able to say exactly which are more motivating than others in their purchase decision. A conjoint’s market simulation approach forces respondents to make trade-offs in the same way they might when making real-world decisions. Even if consumers are unaware of which attributes sway their decision, a conjoint analysis will reveal them. Some common business questions that can be answered with conjoint include:
- What product/service configuration maximizes potential market share and/or revenue?
- What price point(s) are ideal for a given configuration?
- Can we increase price without negatively impacting share?
- What additional value can be offered to offset a pricing increase?
- How is share impacted if competitors change their pricing strategy or value props?
The consumer preferences extracted from conjoint can be fed into sales, marketing, and advertising strategies in any business. Product design, product development, product management, branding issues, package design, pricing research, and market segmentation exercises all benefit from conjoint analysis.
Back to Table of Contents
Different types of conjoint analyses
All conjoint studies compare total ‘packages’ of different products/services (the unique combination of attributes that each has), as well as the variations on those different attributes (called ‘attribute levels’ - for example within the attribute of ‘flavor’, the levels might include spearmint, fresh mint, and cool mint). How each attribute and level affect a respondent’s choice is calculated into a numerical value called a preference score (also known as a utility score). This can then be used to model ideal product scenarios, combining motivating attributes with optimum price points to see how they would affect projected market share and/or revenue.
Within this method there are different types of conjoint analyses; three examples are given below.
1. Choice-based conjoint analysis
Also known as discrete choice conjoint analysis, CBC is the most-used form of the method. One of its main benefits is that it reflects a realistic scenario of choosing between products rather than questioning respondents directly about the importance of each attribute. Respondents are shown sets of 3-5 concepts at a time and asked to choose their favorite, then the importance of each attribute is inferred from their choices. This is a powerful way to understand which features are most important to include in a new product and how to price it. From this information, brands can derive optimal product configurations.
2. Adaptive conjoint analysis
Adaptive conjoint analysis is a flexible approach to CBC, adapting as a survey progresses so that the choice sets that each respondent sees depend on the preferences they have expressed up to that point. Tailoring the questions to each individual streamlines the approach to CBC, as it doesn’t waste time showing product concepts that the respondent would not find appealing, thus cutting down the length of the survey. Respondents can also find this type of conjoint experience more engaging when they see the survey is reacting to their personal preferences.
3. Self-explicated conjoint analysis
This version of conjoint analysis takes the focus off the package of attributes as a whole and instead zooms in on the attributes and attribute levels. Respondents are able to eliminate attributes they wouldn’t consider at all, as well as choose their favorite and least favorite attribute levels. The remaining levels are then rated against the most/least favorite level. The importance of the favorite attribute in the context of the product as a whole is calculated, and a utility score is given for each attribute and attribute level. This CBC method doesn't demand the same level of statistical analysis as other types of conjoints, but it isn’t best suited to pricing research as price can’t be fairly compared to other attributes.
How to create a conjoint analysis survey, with example questions
In product research
Suppose a fast food chain wants to introduce a new burger to its menu. Some of its business questions before launching this new menu item might be around ingredients, calorie content, taste, and price. The new burger has a number of potential options that it could offer to the market but the fast food business needs to be sure that the options it chooses will have optimal uptake among their diners. Using conjoint analysis, the fast food chain could test the following combinations of product features to respondents, asking them to identify their preferred burger:
Option 1: Plant-based, spicy, 650 calories, $5.25
Option 2: Prime beef, spicy, 790 calories, $4.29
Option 3: Prime beef, with cheese, 820 calories, $4.99
Option 4: Plant-based, with cheese, 900 calories, $6.19
Rather than asking respondents about ingredients, taste, calories, and price separately, conjoint analysis presents the features within a realistic product context and analyzes the results to reveal which features are the most powerful in driving purchase.
In service research
Similarly, service propositions can be presented with a variety of feature combinations for respondents to choose from. For example, a TV & broadband service set of options might look like this:
Option 1: Fast broadband, basic TV bundle plus sports channels, $50 per month
Option 2: Superfast broadband, basic TV bundle, and no sports channels, $90 per month
Option 3: Superfast broadband, TV with movies and sports channels, $120 per month
Option 4. Fast broadband, TV with movies but no sports channels, $100 per month
Conjoint analysis will reveal which parts of the service respondents attach the most importance to and which they will pay more for.
Back to Table of Contents
How does conjoint analysis help interpret preferences?
Consumers are faced with product decisions every day and there's a lot of different aspects that go into choosing a certain product over another - but often, we don't know what those are. Many standard usage and attitude questions don't quite get into the nitty-gritty details of consumer decision making that businesses need to succeed. Conjoint analysis (a type of advanced methodology) helps researchers unravel consumers' complex decision-making by effectively breaking down product offerings into attributes and levels.
By breaking down the overwhelming amount of products (and their features) into distinct attributes, businesses can hone in on what specific features of a product actually impact final decision making. Researchers can present respondents with different attributes in various combinations to understand the overall preference (and therefore, value) of each individual feature.
Through conjoint analysis, researchers can quantify trade-offs consumers are willing to make between different attributes, providing insights into their underlying preferences. Knowing this, a business is then in a position to use the most preferred features during the final product design phase, to develop a solid pricing strategy, to craft an influential marketing campaign, and ultimately, to enjoy a successful final product launch.
Alternatives to conjoint analysis
As mentioned above, the alternative to using a conjoint analysis for understanding decision making might be left to standard usage and attitude questions such as 'Which product do you prefer' or 'How much do you like this product'?; as you might guess, these outputs aren't going to give you nearly the same actionable insights as a conjoint analysis will.
While conjoint analysis is one of the best methodologies a brand can use to understand feature importance, there are some alternatives for those who don't have access to conjoint or who prefer to go another route:
MaxDiff (Maximum Difference Scaling):
MaxDiff is similar to conjoint analysis in that it forces respondents to make tradeoffs, but the difference is that it focuses on the most and least preferred features individually, rather than evaluating the features as part of set combinations (which is more realistic to a true shoppers' experience).
Van Westendorp Price Sensitivity Meter (PSM):
For brands looking to make a decision around pricing specifically, Van Westendorp is a great method to use to understand how price sensitive your consumers are. Respondents are asked a series of questions to determine acceptable price ranges for a given product (i.e at which price point the product is considered too cheap, too expensive, a bargain, or justified). The intersection of all these price ranges helps identify an optimal price range. But remember, pricing is just one aspect that a brand could test in a conjoint analysis, creating a much more cohesive understanding of a product rather than looking at price alone.
Key Driver Analysis (KDA):
The end goal for researchers running a conjoint analysis is really to learn what's driving consumer purchases. Another advanced methodology that's effective in uncovering this kind of information is, as it's aptly named, the Key Driver Analysis (KDA). A KDA identifies the overall factors or variables that have a significant impact on a particular outcome - such as product purchase intent in this case. This is a useful method when looking to understand the overall impact a variable might have on consumers' decisions, while a conjoint analysis is better at understanding and optimizing individual features.
How to run a conjoint analysis study with quantilope
quantilope’s advanced, automated conjoint analysis method identifies the relative importance of product attributes and attribute levels in a category to help you create a product that offers the optimal configuration of those attributes. The market simulator will predict how market share would change as a result of lowering or increasing the price, or by tweaking any other attributes that make up the product. It can also show how preferences vary by customer segment.
All you have to do is decide which attributes you would like respondents to trade off against each other. Once you have your set of attributes, you simply need to drag the pre-programmed method into your survey before it goes live. Upon survey data capture, quantilope’s AI-driven tools analyze the data using statistical techniques.
Check out quantilope’s conjoint analysis approach to pricing research for an example of how this method can be applied for this particular use case. This brief demo video also shows how straightforward it is to set up a conjoint study on quantilope's platform, as well as how simple it is to interact with findings and create the optimal product profile - in this case, an energy drink. The video demonstrates how altering the price can affect market share, and how changing other aspects such as ingredients and packaging can affect potential uptake.
To learn more about how your brand can leverage quantilope's conjoint analysis for your own brand needs, get in touch with us below:
Request a demo!