SightX Survey Research Guide: Conjoint Analysis Explained

Customer Satisfaction Surveys (CSAT).
Data analytics
Survey analysis
Advanced statistical techniques
Survey research
Survey software
Joint analysis
Agile quantitative research

If you work in or around market research, you’ve most likely used or (at least) heard of blended techniques.

We will be the first to admit that the literature on this topic is often intense and can seem overwhelming to those unfamiliar with the details of conjoint analysis.

But that doesn’t have to be the case.

While it is SightX A platform can automate your curiosity and research projects, it is still crucial to understand the methodologies you are using.

To save you from falling into a technical white paper, we’ve collected some of the most common questions we get about conjoint analysis.

Stay tuned

Subscribe to receive Research Tool Radar and essential updates via email from Insight Platforms.

What is conjoint analysis?

The concept itself is quite simple.

Conjoint analysis is a market research approach that measures the value consumers place on features of a product or service. It does this by uncovering the rules that consumers explicitly (and implicitly) use to make their purchase decisions by mimicking real-world trade-offs when making a purchase.

History of joint analysis

Now a little background story.

Conjoint analysis can be traced back to the 1960s, and was created by mathematical psychologists and statisticians Luce and Tukey (1964). The two have published an article in which they explore how measuring the goodness of specific characteristics of an object can make it possible to measure the goodness of the object as a whole.

Professor Paul Green would later realize the marketing implications of this work; believing it could help marketers understand how customers make complex purchasing decisions and ultimately predict customer behavior.

This would eventually lead to Green co-authoring a landmark article with Vithal Rao that detailed the first consumer-oriented approach to the methodology.

Over the decades since, conjoint analysis has evolved and become increasingly popular in the marketing research industry.

What is the purpose of joint analysis?

Conjoint analysis has many purposes, but it is most commonly used to discover consumer preferences about your product to predict acceptance, assess price sensitivity, select the optimal feature set, and project market share.

The most popular use of conjoint analytics is generally within product development, helping brands find the perfect set of product features and messaging points for their target market.

Conjoint analysis is an incredibly versatile methodology and can be used in most industries for products ranging from travel arrangements to consumer packaged goods.

When should I consider conducting a joint analysis?

When making a purchase, consumers often take into account different product characteristics

These features can be called “attributes” of the product, where each “attribute” has several levels or options.

To illustrate, let’s use an example: a large car manufacturer is considering adding a new car to its range. Although they have some ideas for a new vehicle, they want to know what a customer would be most willing to buy. So they decided to run a pooled analysis to investigate the three attributes.

The first attribute is the color of the car, which makes the color selection dependent on the attribute levels (black, white, or gray). Another attribute is the price of the car ($25K, $35K, $45K). And the third attribute is the type of energy the car uses (gas, hybrid or electric).

A joint analysis will show the company what appeals to consumers the most, which levels are most popular and which is the optimal combination to increase sales.

Steps for performing conjoint analysis

Performing conjoint analysis is much easier than you think! Just follow these simple steps:

Identify the attributes

The first step is to collect the product attributes you want to test. In this case, “attributes” refer to product features.

While it can be easy to get caught up in the minutiae of your product, remember that less is more here.

Adding too many attributes to your co-experiment will only make it harder for respondents to accurately assess your product’s features. In general, we suggest keeping the number of attributes close to 3.

Set your levels

While “attributes” are the features themselves, “levels” are the options associated with each attribute you test.

Similar to attributes, adding too many levels to your experiment will create a large cognitive load for subjects and may even overwhelm them. Then again, less is more here. We generally do not recommend more than 4 levels per attribute.

Enter your information in SightX

While there are certainly manual methods for creating merged analysis, automation can be a HUGE time saver.

Once you’ve collected your attributes and levels, you can easily create a shared experiment within SightX platform by selecting “Conjoint” from the methodology menu.

You can include an instructional description for respondents, giving them a brief background on the category or types of products they will be evaluating. You can also add pictures to make the process more attractive.

With the information you enter, SightX the platform will generate a balanced experiment.

Choose your sample size and run your experiment

Once your experiment is ready to run, it’s time to consider your sample. It is important to include a sufficient number of subjects in your joint experiments, you can use our handy common sample size calculator to determine the right sample size for your project.

Once you know the sample size needed for your pooled analysis, you can launch your project to begin collecting data.

As for the analysis, we’ll cover that below:

Joint analysis outputs

Once you’ve collected the data, your conjoint analysis chart will look something like this:

The chart above will not only show you the importance of each attribute, but also the popularity of each level within the attribute. Ultimately, this data will help you better understand the optimal feature set for your product.

SightX provides three types of common data analysis: partial value, relative partial value, and importance. You can switch between the three on your chart.

What is partial value and how does it relate to conjoint analysis?

While several valuation models are listed above, the most commonly used is the Part-Worth model.

Unlike other models, it makes no prior assumptions about the utility caused by a particular level of any attribution. Simply put, the results will be a more accurate representation of consumer preferences.

In product research, you will see that multiple attributes come together to define the overall value of a product. Some may be more important to consumers than others.

Partial value is an estimate of the total value (or utility) associated with each attribute and level used to define your product. Thus, the values of each separate attribute are partial values.

There are several research techniques for partial value estimation. Latent class analysis and regression modeling based on Hierarchical Bayesian (HB) are mostly used.

The values of partial value utilities provide information about how attractive the attribute levels are. Do you remember examples of car types in different price categories and colors?

If you want to know the relative importance of each attribute, you will need to calculate attribute importance by determining how much difference each attribute can make to the overall utility of a particular product. That difference is the range of attribute utility values.

You can calculate percentages from relative ranges, getting a set of attribute importance values that add up to 100. The higher the percentage, the more important the feature.

Finally, you can discern which features should be combined for maximum effect- ideallyyou work with a software platform that automates this process for you.

What is the relationship between joint utility scores and market share?

In addition to these insights, you may be interested in your utility scores and their use for market share simulation, where market simulation provides information on the relative proportion of respondents who prefer predefined products in a given context.

A market share simulation allows researchers to test different scenarios and assess factors such as price demand curves, the impact of product adjustments, and the competitive environment.

The first step in conducting a market simulation begins with determining the relevant products. The total utility of these products is calculated at the individual or target level – the total utility of a product is the sum of its partial utilities (Green and Krieger, 1988). From there, consumer insights managers can compare the product’s overall utility value to that of the “none of the above” option.

The greater the difference between the total utility rating of the alternative and the utility rating of the “none of the above” option, the more likely users are to accept the alternative. Conversely, if the product’s overall usefulness rating is below the “none of the above” option, this means that users are unlikely to accept the offer.

Researchers can apply the logit model to estimate market share. Market share is estimated by simply powering the total utility and then dividing this value by the sum of the powered values of all products and the “none of the above” option. Here is the formula:

If, like most people, formula milk scares you a little, fear not!

Conjoint analysis is simply a mathematical representation of what we covered in the paragraph above: “Market share is calculated by powering the total utility of a product and then dividing that value by the sum of the power values of all products and “none of the above “options.”

The calculation can be done using a basic calculator equipped with an “EXP” (or ex) button.

Below is an example with total utility scores, exponential scores, and market shares associated with it.

While the above overview should prove useful, we understand that these concepts can be technically demanding.

SightX allows you to automate conjoint analysis, helping you to more easily optimize product development and predict the likelihood of market acceptance. Our step-by-step setup allows you to launch projects within minutes and instantly analyze the results with real-time analytics.

Companies using conjoint analysis

Conjoint analysis is an incredibly popular tool for organizations of all shapes and sizes. Some recognizable names include:

NBC Universal Parks & Resorts used conjoint analysis to design theme park experiences.
Proctor & Gamble relied on conjoint analysis to provide insight into the messaging, pricing and design of its CPG products.
Apple used joint analysis to estimate the economic cost of patent infringement by their competitor Samsung.
Bose used conjoint analysis in its product development cycle and in expanding its product lines.