Our dedicated Analytics team, led by Jon Godin and Sadi Ersace, provides internal expertise and analysis on most of our trade-off based (Discrete Choice, MaxDiff Scaling, Conjoint) and complex engagements. Working with primary data or behavioral data, this ensures that we create effective analysis plans and use the best techniques to find the answers you need.
Part of Jon’s role is to ensure that the company has access to and training on the most useful set of tools available for analysis and data mining.
Some of our current favorites:
Adaptive Choice Based Conjoint (ACBC)
ACBC is a conjoint design whose utilities produce higher hit rates and lower mean absolute errors than standard choice utilities when used to predict responses to holdout tasks, especially situations with small samples of respondents, so we are essentially getting more information about each respondent’s preferences using than with other approaches.
MARS is a sophisticated stepwise, piecewise regression model that includes only the variables that really matter, and explains relationships in the data with less bias and more accurate predictions than standard regression models.
TreeNet is a data mining technique that allows us to turn an excessive amount of data into a very accurate predictive model.
Experience an Anchored MaxDiff design with our new Ice Cream Flavor rater MaxDiff demo
See how Adaptive Choice Based Conjoint (ACBC) works firsthand with our new Burrito Builder ACBC demo
Experience ACBC from a respondent’s perspective by trying out our Build-Your-Own Digital Camera ACBC demo.