AI Copilots for Conjoint Analysis
Custom extensions to the traditional outputs from a conjoint activity (e.g., share of preference) are common and the time spent developing code to accommodate this stage of analysis is cumbersome.
We are now able to implement AI copilots to assist our team when developing these solutions, cutting down on coding time while optimizing code efficiency. This is a new and evolving space, with incremental value to be unlocked moving forward.
While analyzing the results from a recent conjoint study, our Decision Sciences & Innovation team identified key improvement areas within the code to synthesize traditional conjoint outputs with data from a separate stated activity in the survey. The team leveraged an AI copilot to build several custom functions that would streamline the analysis, combining different types of data into a single cohesive output.
This was a strong start, but we didn’t stop there. Next, we fine-tuned the prompts to optimize the functions, maintaining the desired output with substantial enhancements to speed for greater efficiency.
Our team of data scientists harnessed the power of generative AI leveraging this sophisticated tool to enhance the way traditional problems are solved while simultaneously reducing time to insights. Integrating AI into current workflows will save countless hours of development time, freeing up teams to focus more on creative problem solving.
