Node Description

Simulate Conjoint Node

Conjoint Analysis is a market research tool that helps determine how Customers value different Products Attributes. Product Attributes are those features, functions, and benefits (including Price) that characterize Products. Conjoint Analysis is used in marketing, product management, and operations research. It is frequently used to test Customer acceptance of innovative new Products and Services, evaluate Product Positioning, and test new Pricing.

Conjoint Analysis traditionally depended upon physical surveys that caused test subjects to make trade-off decisions between a set of Product Attributes. For example, the test subject may be asked whether they would be willing to pay a specific premium for a higher performance version of a given Product. The survey format presents each test subject with a choice between at least two Products, along with a subset of Features from which to base a selection.

This ‘Simulate Conjoint’ node runs a Conjoint Analysis using data from a pre-calculated Willingness To Pay Matrix (Input WTP Matrix). The WTP Matrix can be generated through a combination of upstream ‘Differentation Horizontal’ and ‘Differentiation Vertical’ nodes used to model each of the Product Features found in the Market.

While these upstream modeling nodes provide precise details about the part-worth values that Customers have for Features, it would be a mistake to equate part-worth values with Feature Importance. A Feature may provide tremendous value to a Customer, and yet be unimportant to whether the Customer purchased the Product. For example, a Customer may decide to buy a car because of the value it provides in getting them around. But this fact alone would be irrelevant to whether the Customer buys a Ford or a Toyota. As all cars provide the benefit of ‘transportation’ this becomes an undifferentiated and irrelevant Feature. Instead, the Customer may place the greatest important on ‘color’ when deciding which car to purchase.

The results of this Conjoint Analysis can help during New Product Development cycles. New Product concepts can be designed to fill gaps in a Product Line or satisfy the unmet needs of emerging Customer Segments. Market Simulation can be used to predict future Market Share and optimize Prices.

This Premium Node is not available as part of the free Community Edition. Premium Nodes help clean and connect real-world data to Market Simulations, and provide advanced Market Science analysis. Note that these descriptions are often deliberately vague.


Simulate Conjoint

The Simulate Conjoint node runs a Conjoint Analysis using data from a Willingness To Pay Matrix. The Input WTP Matrix can be generated upstream with ‘Differentation Horizontal’ and ‘Differentiation Vertical’ nodes. The Simulate Conjoint node calculates the Customer Importance of each Attribute.


Product Array

The complete set of all Products found in the ‘Input Product Attributes’ table and the ‘Input WTP Matrix’.

Product Attributes

Lists the set of Categorical Attributes that are found within each Product. Categorical Attributes can be listed in either this table or in the ‘Input Product Array’ (Numerical Attributes can only be found in the ‘Input Product Array’).

WTP Matrix

The Willingness To Pay (WTP) Customer Distribution matrix for each Product column in the Market by each Virtual Customer row.



The user selects all of the Attributes to include in the Conjoint Analysis from the Input Product Array. Columns with String and Boolean values will be treated as categorical Attribute-Levels. Columns with Double and Integer values (such as Price) will be treated as continuous Numerical Attribute.


Product Rankings

The Output Product Rankings lists the calculated results from the first step of Conjoint Analysis. The estimated preference for each Product is ranked by each Customer according to their ‘Consumer Surplus’.

Partworth Utilities

The Output Partworth Utilities table contains the calculated results from the second step of Conjoint Analysis when a multiple regression analysis compares each Customer’s Total Utility against the presence or absence of the Categorical Attributes of the Products, as well as the Numerical Attributes of the Products. The beta coefficients are calculated for each Attribute-Level according to a Least Squares Multiple Linear Regression algorithm.

Attribute Importance

 The Output Attribute Importance table contains the results from the final step of Conjoint Analysis when Part-Worth Utilities from individual Customers are aggregated into overall Importance Levels.