How to run a Gabor Granger analysis

The user types something like:

• "What price maximises revenue for this product?"
• "Run Gabor Granger on our pricing data."
• "How elastic is demand at the prices we tested?"

The platform recognises the intent and selects Gabor Granger automatically.

The chat asks the user to confirm which column contains the maximum willingness-to-pay values. Unlike Van Westendorp (which needs four columns), Gabor Granger needs just one:

• Maximum willingness to pay

The platform suggests a mapping if a column name looks like a match (e.g. max_wtp, willingness_to_pay). Otherwise it asks the user to pick.

The chat asks whether the user wants to filter the data:

• "Only premium subscribers"
• "Exclude respondents who said N/A to the pricing question"

The platform validates that enough responses remain after filtering.

The chat asks whether the user wants to split the analysis:

• "Compare loyal customers vs. new prospects"
• "Split by distribution channel"

If set, each group gets its own demand curve and revenue-optimal price.

The chat shows a final confirmation screen:

• The mapped WTP column
• Any active filters
• The subgroup variable (if set)
• The expected sample size

The user confirms, and the analysis runs.

Within seconds, the chat returns:

• A demand curve chart — price vs. acceptance rate
• A per-price-point table — price, demand %, respondent count, projected revenue at each tested price
• The revenue-optimal price — highlighted in the table and chart
• Summary statistics — n, price range, max revenue
• A written summary calling out the optimal price and key observations

If subgroups were requested, each group gets its own set of results plus a comparison highlighting where pricing power differs.

The user can refine in the same conversation:

• "What if we exclude the outlier prices above $200?"
• "Split that by age group instead"
• "Now run Van Westendorp on the same data to see the acceptable range"