The data analysis identified three higher-level success factors with 14 questions that showed a statistically significant correlation with the purchase decision.
An AI-based forecasting model that uses gut feeling and success factors to predict the probability of success. Receiver operating characteristic (ROC) curves measured the accuracy of the predictions. The area under the curve (AUC) is a commonly accepted indicator of the accuracy of the predictions.
By comparing the AUCs of the forecasting model with the gut feeling (see figure on the right-hand side), it was possible to demonstrate empirically for the first time that the predictions of the forecasting model are significantly more accurate than the gut feeling of the salespeople.
According to the classification, the ROC showed a fair prediction quality (AUC value: 0.76) for the gut feeling and an good prediction quality (AUC value: 0.84) for the forecasting model.
The hit rate of the AI-based forecasting model reached an impressive value of 82%.
That means that the forecasting model predicted the correct buying decision in 82% of all predictions, regardless of whether salespeople won or lost their opportunities.