Latest
Choice Modelling
6 min read
Using a Market Simulator: Price, Share and SKU Line-ups
A conjoint study earns its keep in the simulator. How share-of-preference rules work, how to read price curves and elasticity, how WTP is calculated, and how to build a line-up that reaches the most customers.
How Many Respondents Do You Need for Conjoint Analysis?
There is no single right sample size for conjoint, but there are well-tested rules. Johnson's rule sets the floor, 300 is the usual minimum for robust work, and every group you want to compare needs about 200 of its own.
CBC vs ACBC: Which Conjoint Should You Run?
Choice-based conjoint is the default; adaptive CBC earns its longer interview when a product has many attributes. What each method asks, what the evidence says, and how Ordeen chooses between them.
What is Conjoint Analysis? Methodology, History and Use Cases
Conjoint analysis measures how customers trade off attributes of a product. The gold standard for pricing and portfolio decisions.
Ordeen Choice Modelling — Frequently Asked Questions
Answers to 20+ questions about methodology, pricing, panels, data residency, integrations, support, and migration from Sawtooth.
Ordeen Choice Modelling — Full Statistical Methodology
The full statistical specification of Ordeen’s choice modelling stack — hierarchical Bayes, HMC/NUTS, Krinsky-Robb WTP, BH-corrected interactions.
What is Hierarchical Bayesian Estimation in Choice Modelling?
Hierarchical Bayes extracts individual-level utilities from the sparse per-respondent data of a conjoint study — the gold standard.
Load More
