Electric Insights
CPG Demo · Beer Concept Test (n=803)

Model Builder

Find the drivers of purchase intent

Build a logistic regression of purchase intent for the test product. Pick predictor variables — demographics, beer-drinking habits, brand awareness, package reactions — and see which ones move the headline.

The outcome is a top-2-box recode of Q21: 1 if a respondent rated purchase intent Extremely or Very likely; 0 otherwise. Baseline rate is 54.7% in the unmodified sample.

How to Use This Tool

The published model is loaded — six package-reaction questions are pre-selected. Click Run Analysis at the bottom to fit it, or add and remove variables to build your own version.

1. Outcome

The outcome is fixed for this case study: Purchase Intent (Top-2-Box on Q21). 1 = Extremely or Very likely; 0 = Somewhat / Slightly / Not at all likely.

2. Pick predictors

Six package-reaction questions are pre-selected — the published model. Use the search box and category buttons (Package Reaction, Brand Knowledge, Beer Preferences, Behavior, Demographics) to find others to add or to swap in.

3. Fit and read

Click Run Analysis to fit the model with whatever variables you've checked. Or use Auto-Build Standard to let the algorithm search all variables automatically, or Auto-Build Actionable Predictors to weight selection toward levers you can move through package design or marketing over fixed context like demographics.

Survey Questions to Include in Your Model

The six highlighted variables are from the published purchase-intent model. Check additional variables to test whether they improve how well the model predicts the outcome, then click Run Analysis. Or use Auto-Build Standard to let the algorithm search all variables automatically, or Auto-Build Actionable Predictors to constrain selection to independent, movable levers.
Loading variables

What are you trying to predict?

Choose the outcome your model will try to explain.

Required
Loading outcomes…
r =
-0.8 0.8
r =
0 0.8
2 (Binary) 5 (Max)
Modeling:
0 of 20 variables selected
Run Analysis

Build a model from exactly the variables you selected. Full analyst control.

Auto-Build

Search all variables automatically. Choose a selection strategy:

Best predictive fit across all available variables

Auto-Building Optimal Model… 0s
  • Initialising…
  • Screening candidate predictors
  • Running forward stepwise selection
  • Fitting final model
  • Computing diagnostics & margins

Section 1: Full Model Performance

Complete model including all predictor variables and the synthetic variable