Learn how to use conjoint analysis for market research and content strategy.
When you’re developing a marketing strategy, one of the first and most important steps is to do your research. In order to actually sell your product or service, you need to fully understand your target audience when they’re in an actual buying situation:
- What gets their attention?
- Why would they choose you over a competitor?
- How do they make their purchase decisions?
One of the most effective ways to answer each of these questions is with conjoint analysis.
Defining conjoint analysis
Conjoint analysis is a survey-based statistical analysis technique used during market research that quantifies the value customers place on attributes of a product or service.
Already that may sound complicated, so let’s break it down a little more.
Conjoint analysis is a statistical technique that uses a survey to determine consumer preferences before they make purchase decisions. It asks each respondent a series of questions—also known as choice tasks—in which they select between a few packaged options based on relative importance and what they deem most valuable.
Each package presents different product features, called attributes, and each attribute shows multiple types, or levels. Here are some examples of attributes:
- Product material
- Price
- Time for completed service
- Distance or location
- Company/brand features
Next, here are examples of each attribute level:
- Product material: leather, polyester, denim
- Price: $12.99, $15.49, $19.29
- Time for completed service: 7 minutes, 15 minutes, 25 minutes
- Distance/location: 2-minute walk, 5-minute walk, 10-minute walk
- Company/brand features: Women-owned, dermatologist-approved, sustainably made
As they answer each question, respondents determine which trade-off feels like the best deal for them.
Finally, companies use the data from that survey to determine a utility score, revealing which attribute(s) respondents find most valuable. Survey data can also be used to measure the consumers’ overall preference scores, which state a consumer's likelihood to purchase a packaged option based on preference.
Types of conjoint analysis
The reason why this method is called conjoint analysis is because survey respondents have to choose a conjoined product package with different attributes and levels. When developing a conjoint analysis survey, companies can use a variety of statistical techniques. Each method asks the respondent a different series of questions that businesses can leverage for different insights.
Choice-based conjoint analysis (aka discrete choice conjoint analysis)
This is the most prevalent type of conjoint analysis that market researchers use. Choice-based conjoint analysis (CBC)—or discrete choice conjoint analysis—simulates the market and demonstrates how respondents value certain attribute levels. Since this method is also most commonly used to explain how conjoint analysis works, let’s go over an example.
Say you’ve started a car cleaning service and you want to develop a survey that asks customers what sort of cleaning package they’d prefer.
You break down your service into attributes and levels. The attributes could be price, services provided, time spent cleaning, and a manual vs automated car wash. Then you include different levels of pricing, services, and time.
Once your survey is complete and you analyze your conjoint data, you may learn that more respondents would prefer a car service that’s quick and inexpensive. Or you may learn that respondents would rather spend the extra $10 for a wax and tire polish—despite the service taking much longer.
Adaptive conjoint analysis (ACA)
Adaptive conjoint analysis (ACA) is similar to CBC analysis. The difference though, is that each question updates in real time and adapts to each respondent's choices. Adaptive conjoint analysis is ideal for when respondents need to evaluate more attributes than in a choice-based survey.
This way, companies can get a full perspective of what their customers value and are looking for by presenting combinations of attributes and levels that companies may not have thought of before. ACA is also a more efficient way of surveying respondents because each follow-up question becomes more curated to each input answer, which makes the survey feel more relevant and pertinent to the respondent’s values and desires.
Full-profile conjoint analysis
This conjoint analysis technique requires a full description of each product in a choice task. Other market research techniques usually limit the number of attributes, but with full-profile conjoint analysis, the respondent is able to see a thorough description with every attribute. Respondents then select which product they’d purchase with maximum likelihood.
Menu-based conjoint analysis
Typically, a conjoint survey doesn’t ask respondents outright what they’d like to pay for a product or what features they’d like to see with it. Menu-based conjoint analysis surveys differ because they enable the respondent to package a product by themselves. This allows companies to see how potential customers may value certain combinations of attributes and levels.
Of course, everyone would like to have the best-quality product that’s inexpensive, comes with all sorts of benefits and features, or takes the least amount of time to finish or arrive. However, menu-based conjoint analysis surveys prompt respondents to categorize each predetermined attribute and level so they can customize a packaged product that they feel would deliver the most value.
Why use conjoint analysis for market research?
Companies often conduct conjoint analysis surveys because they are one of the best survey methods for determining customer values and preferences during the buying process. Let’s go over the business benefits of conjoint analysis and why it’s so effective.
Highlight consumer preferences
When companies know which product features are the most valuable to consumers, they can highlight them in their advertisements. Say, for example, you learn that one respondent group values your brand’s environmental mission and another group values the quality of your materials. With data from your conjoint study, you can target some consumers with ads that highlight your stance on climate change while other ads target consumers that are looking for a brand with high-quality materials.
Companies can also use a conjoint analysis experiment to determine which new product features to add or take away based on survey data, utility scores, and preference scores. If you learn that most respondents preferred an old feature compared to a potential new one, you could save time, money, and resources that would be spent launching new products with multiple features that your customers wouldn’t prefer as much.
Mimic real-life trade-offs
People make trade-offs every day. However, not all trade-offs are created equal because different people have different priorities. For example, some people trade sleeping in so they can go to the gym and grab breakfast to go before starting work; others may prefer to sleep in more and prepare a quick breakfast at home before work.
Conjoint analysis mimics this kind of daily trade-off. For instance, when it comes to purchase decisions, consumers often trade:
- Higher or lower price for quality
- Timeliness of a service for the amount of available services
- Imported items for locally made goods
As mentioned, conjoint analysis doesn’t necessarily ask respondents what they specifically prefer in a product or service. Instead, it demonstrates a realistic context by asking respondents to choose which packaged option they prefer, ultimately revealing which attribute level respondents are willing to trade for another.
Develop insightful product and pricing research
Conjoint analysis methods allow companies to gain insights on how much a consumer monetarily values their product or service. By developing conjoint surveys that focus primarily on product and pricing research, companies can understand how much consumers are willing to pay.
Simulate competitive markets
Businesses can develop surveys that employ a brand price trade-off approach, wherein they learn if consumers have a bias toward a competitor solely based on a name brand. This allows companies to simulate a competitive market situation, allowing them to see whether or not customers would prefer them over another brand and why.
Predict marketing trends
Instead of hoping that a new product, feature, or service will land well with new consumers, conjoint analysis can help companies make more informed decisions with their marketing strategies. Companies often use conjoint analysis to forecast potential demand, predict marketing trends, or determine product acceptance before they launch by noticing trends and quickly acting on relevant data.
5 steps to creating a conjoint analysis survey
Let’s go over the 5 steps to creating an effective marketing research campaign with a conjoint analysis survey.
Step 1: Define attributes, levels, and a pricing structure
When you start developing a conjoint survey, make sure you can succinctly define each attribute and level, and that you have enough to present a few packaged options to a respondent. Break down your product or service into attributes of interest and define each level that you want to evaluate. Take a look back at our choice-based conjoint analysis example for ideas of where to start.
Also keep in mind what your price range is so you can position each price level as slightly above and slightly below your range. This ensures that respondents don’t all reply that they’d want a price that’s much lower than you’d like.
Note that the only way to move forward from here is if you have succinct, well-defined attributes, levels, and pricing. If these aren’t defined, you likely won’t get a lot of actionable insights.
Step 2: Create a conjoint survey
There are tons of websites that provide conjoint analysis software, and some even allow users to create a survey for free. You can also create a survey on platforms like SurveyMonkey or with any of Mailchimp’s free survey tools.
Before you start, think about which conjoint analysis method you’d like to use. Here’s a quick breakdown on how to use each type of conjoint analysis we covered:
- Choice-based conjoint (CBC) analysis/discrete choice conjoint analysis: Asks respondents to choose which packaged option they would most likely purchase.
- Adaptive conjoint analysis (ACA): Ideal for a large number of attributes; adapts to each respondent's choices and develops new questions in real time.
- Full-profile conjoint analysis: Presents respondents with multiple full product descriptions, prompting them to choose the product they’re most inclined to purchase.
- Menu-based conjoint analysis: Enables respondents to package a product or service themselves based on their preferences.
Step 3: Invite survey respondents
You invite survey respondents by sending an email to subscribed customers or potential consumers. Always let your invitees know that their opinion matters and that your survey is for consumer preference research that will help you better understand them. You can even provide a small incentive for participating in your survey, such as a discount or the chance to win a gift.
However, before you send out survey invitations, make sure you have reliable resources to collect their response data. Most survey platforms offer ways to manage invitations and collect insights. However, some offer more helpful analysis insights than others, so make sure to choose your survey platform wisely.
Step 4: Analyze survey data
Once the responses to your market research survey are ready, your selected conjoint software or survey tool will analyze the data and provide insights on each utility and preference scores. Each score will outline which product or service respondents are most likely to purchase, which features are most desired, and which attributes have the most impact during the buying process.
Step 5: Deploy conjoint data in marketing
Finally, you have your data, preference scores, and other valuable analysis insights. However, remember that it’s okay to not get everything you need after one survey. You may need to do a few rounds of surveys for different audiences or at different times. It’s possible you may have to adjust your survey completely if you find that another conjoint analysis technique may be more helpful. Nonetheless, your first survey can help you start implementing your newfound understanding into marketing strategies, campaigns, and advertisements.
Also, no matter what you learn, just know that it can take weeks or months to implement your findings into a marketing strategy. Always set realistic expectations and stick to a schedule in order to stay on top of your marketing goals.
Understand the consumer with conjoint analysis
Conjoint analysis is an ideal way for businesses to learn more about trending preferences and pivot upon learning what consumers like or don’t like. Deploying a conjoint analysis study not only enables businesses to effectively provide what consumers are looking for, but it can save businesses time, money, and resources that might get squandered by just guessing what is currently trending.
Even if your findings surprise you and your team, a conjoint analysis method can help you strengthen your marketing strategy and get you closer to your customer base—which is really valuable in the long run.