Discrete Choice Experiments


Nik Samoylov and Ujwal Kayande have developed a tool for choice-based conjoint of potential use in Master's classes

Dear Colleagues:

Nik Samoylov, my former student and ex-consultant at Bain, and I have developed an online tool to do discrete choice experiments (commonly called choice-based conjoint). The tool – available at www.conjoint.online – can be used without charge for teaching and academic research purposes.

The tool:

  1. allows users to design generic or alternative-specific designs using d-optimality as the criterion,
  2. creates an online interface to collect data,
  3. allows users to send an email link to respondents (or buy panel respondents from global research panel providers), and
  4. analyzes the data using hierarchical bayes techniques, producing individual level partworths.

Users can also conduct a post-hoc segmentation on the individual partworths, profile customer segments by demographics, estimate marginal willingness to pay if price is included as an attribute, and conduct market share simulations (also by segment). Users can also download a PowerPoint file with graphs, as well as an Excel sheet for profitability analysis if margins per attribute level are known.

In MBA and Executive MBA classes, I first help students develop a conceptual understanding of conjoint/discrete choice, demonstrate the tool in class, and then provide access to the tool for project work. In Executive Education, I first describe conjoint, and then do a live demonstration of the tool by getting them to design a study, collect and analyze data with participants as the respondents, and then spend most of the time on business insights provided by the analysis.

The tool has also been used commercially by global multinationals as well as start-ups. To use the tool for teaching or academic research purposes, please request a fee waiver at https://run.conjoint.ly/support (copy and paste the URL into a browser or click on it).