Conjoint Analysis: a comprehensive practical guide

1. Introduction

Developing a new product and establishing it on the market provides challenges for many companies. During the developing process the central questions are: Which features do my customers expect and which of those features are considered the most important? 

Roughly 95% of all new product launches fail. Mostly because the products do not meet the customers’ requirements and expectations nor fulfil their needs. Hence, it is important to do some research to answer the questions above. 


However, these questions can already be answered during the development phase by means of a conjoint analysis. Thanks to this, it is possible to predict the behaviour of potential consumers in advance by presenting realistic purchase scenarios to identify gaps in the existing product competition. 

The word conjoint is a portmanteau of the terms "considered" and "jointly", from which the conjoint analysis also gets its definition: various product features (attributes) are considered together and then weighed against other variants.


2. What is the conjoint analysis?

Conjoint analysis has its origins in psychology and was developed by Robert Luce and John Tukey in 1964. Since then, it has been used primarily in market research and product development to determine at the development stage which attributes consumers want and which they perceive as particularly important. 

Attributes include, for example, certain functions, designs or functional features such as weight, size and price. 

However, since consumers tend to want as many attributes as possible for as little cost as necessary, conjoint analysis takes a different approach than, for example, the MaxDiff Method.

This is because the distinctive feature of the conjoint method is the combination of different attributes instead of independent comparison. 



Both for high-priced products, such as automobiles; hardware, such as laptops or smartphones; and luxury goods, as well as for everyday products or during the conception phase, conducting a conjoint analysis is a good idea and ensures clear conditions.

The concept of the analysis is basically quite simple. In the context of conjoint analysis, consumers are shown different products, each of which differs in terms of the combination of features. In this way, a realistic experience can be created that comes very close to an everyday purchase decision.


Conjoint example in the Appinio app


In this sample conjoint analysis, the aim is to determine which types of chocolate consumers prefer and what price they are willing to pay for each type.

The respective attributes are leveled, i.e. they are displayed in a certain form. For the chocolate example the filling attribute is divided into the levels vanilla cream, strawberries & cream and, kiwi ganache.


In this way, a ranking can be created that shows which attributes are most important and which characteristics are most attractive. 

This evaluation can then be used to decide which combination is both most appealing to consumers and most profitable for you and your company.


Evaluation example of a conjoint analysis


3. What is the difference between the conjoint method and Discrete Choice?

While there are some similarities between the conjoint analysis and the Discrete Choice Model (short: DCM) there also are some distinctive differences. Both models are preference structured models that are designed to see what's underlying a consumption choice. The big difference: While respondents see each of the product profiles with their respective attributes in smaller groups, in a DCM they will see all of the products simultaneously. 


Therefore DCM is a bit more realistic when it comes to predict buyer behaviour than the conjoint analysis. On the other hand the DCM can be overwhelming for the respondents as they see plenty of options. With a conjoint analysis it is also possible to gain more information about the attributes' relativity and importance to each other and their contribution to he final buying decision.


Another advantage: the conjoint analysis is a great way to predict behaviour before the product is launched. This is rather unlikely when using a DCM. 


4. The Choice-Based conjoint method

Choice-based conjoint analysis (CBC for short) is the most frequently used form of conjoint analysis. And not without reason. This is because this type of conjoint analysis deliberately aims to get consumers to decide between variants and thus to accept trade-offs.


The choice-based method therefore offers a detailed analysis that is nevertheless based on realistic scenarios that are close to the market. After all, everyone makes a large number of decisions every day and weighs up different attributes against each other.


In the Choice-Based conjoint analysis, all previously defined attributes are combined evenly so that a statistically valid ranking can be created at the end of the analysis.


In addition to CBC analysis, which is performed at Appinio, there are other types of conjoint analysis. These include the Adaptive Choice conjoint and the Menu-Based conjoint analysis. However, they cannot be used as flexibly as the Choice-Based conjoint method.


5. Three Examples of use for conjoint analysis

Since individual attributes can be defined for each new execution and combined with each other as desired in a conjoint analysis, this market research method is suitable for a wide range of use cases.


Concept testing

Right at the beginning of the product development stage, it is a good idea to carry out a conjoint analysis. In this way, concepts that are not accepted by consumers can be directly disregarded. Instead, potentials can be identified and then further developed based on the evaluation. In the best case, this procedure can save valuable resources and capacities and also minimises the risk of a failed product.


Diversifications and product range expansion

Even if a mature product range already exists, conjoint analysis is a good choice. Especially in the case of diversifications such as new product sizes, flavours or colours, it can be essential to test these beforehand through market research. 

By the way: The conjoint method can also be used for packaging and design tests.


Price determination with the conjoint method

To perform a price analysis using this methodology, a Van Westendorp price analysis can be performed as a basis. However, the conjoint analysis also works wonderfully as a stand-alone method for price determination. In addition, this variant has the great advantage that several concepts can be tested directly with regard to their willingness to pay a price and not just one product.


6. Conjoint method: These are the best practices

  1. Use short and concise descriptions of product features. This helps to avoid misunderstandings that could possibly distort the analysis.
  2. Use pictures. This makes it easier to distinguish between the different variants and makes it even easier for respondents to imagine what the survey is about.
  3. Use descriptive comparisons for attributes. Everyone perceives abstract attributes such as weight or size differently. Therefore, levels such as "light", "heavy" or "large" and "small" should be avoided. In such cases, concrete comparisons are more appropriate, such as: As heavy as a similar product.


By the way, the best way to implement these tips is with the Appinio Conjoint Analysis Tool, because there you will find the corresponding setting options. Please feel free to send us a non-binding inquiry.


7. Setting up a conjoint analysis (with Appinio)

Step 1:

  • Register on the Appinio platform.

    Define the 3-4 most important product features (e.g. price, design) to be tested.

    Contact one of our market research experts. They will guide you with formulating the definition of the product features right up until your survey goes live.


Step 2:

  • Go live with your survey
    • Our professional market researchers do a final check of your survey before it goes live.
    • Afterwards, your survey will be answered by our panel immediately.


Step 3:

  • Analyse the data:
    • Analyse your base data in our interactive dashboard in real time.

    • The results of the conjoint survey are calculated and visualised in bar graphs and tables by our research consultants to show the utilities and importance of each factor.

    • Accordingly, the results can be used immediately for decision-making.

      Export PowerPoint, Excel or CSV files at any time.

Importance of attributes in relation to each other
Utility of the specific features 

8. What are the advantages and disadvantages of a conjoint analysis?


  • Thanks to the conjoint analysis, it is easy to determine which features a product must have and which the consumers would forgo.
  • Due to many different combinations of attributes and levels even subconscious decisions can be measured and analysed with a conjoint analysis.
  • The research design is highly flexible and can be adapted to almost every product or concept.
  • The conjoint method is incredibly versatile. A multitude of studies can be covered in a single study: price willingness, design tests or product attributes.




In addition to the advantages, as with any method, there are also a few disadvantages in the analysis. For example, it is possible that:
  • Respondents opt for luxury variants because they don’t spend any money and consequently have no sense of making a real purchasing decision.
  • This can lead to a discrepancy between survey results and actual market behavior.

9. Conclusion

Carrying out a conjoint analysis offers a multitude of advantages and can be used for a wide range of use cases. The conjoint method is particularly suitable in the areas of product development and marketing. Another particular advantage is that several combinations and variants can be tested without consumers having to choose their favourites from a list of attributes. The conjoint analysis realistically reflects an everyday purchase decision.

You can call this via showToast(message, { variant: 'normal' | 'error' }) function