Choice modelling

Uncover customer preferences by simulating trade-offs in decision making

OVERVIEW

Choice modelling is a statistical technique that simulates real-world decision-making scenarios to understand how customers make choices based on the various features and attributes of products or services. Broadly speaking, it enables you to determine the relative importance (the technical term is utility) of various attributes and the impact of different levels within each attribute.

There are many types of choice modelling depending on the context of the research question. Some of these include conjoint analysis, MaxDiff and Brand Price Trade Off (BPTO). At Lonergan Research, we employ the technique best suited to your research question and objectives.  

HOW CHOICE MODELLING WORKS

Our choice modelling methodology:

  1. Attribute and Level Identification: We work closely with you to identify the key attributes and levels that define your product or service offering.   
  2. Experimental Design and Data Collection: We design the experiment that presents respondents with choices varying by attribute levels. The systematic varying of attribute levels, allows us to analyse trade-offs that customers make when weighing up decisions.
  3. Statistical Analysis: We employ advanced statistical models (Hierarchical Bayes, multinomial logit, etc) to analyse the collected data and compute preference utilities. These represent the relative importance of each attribute and its respective levels. The output of the choice model can be used as inputs for further statistical analysis such as market segmentation, regression, and driver analysis.
  4. Insights & Recommendations: Using the preference utilities, we provide insights such as optimal combination of product features or pricing.

APPLICATIONS

EXPERTISE REQUIRED

Need expert help? Contact Lonergan Research

Share