Predicting customer satisfaction in real-time
Facing two challenges – understanding who is dissatisfied and then being able to react in a timely manner – Domino’s came to us to look past their data and achieve what was previously impossible: assessing over 1.4 million orders a week and predicting a customer’s satisfaction in real-time.
We built a database bringing together the customer, what was ordered, the production and QA process, the delivery time, and satisfaction survey responses of every single order from the last five years. Building a ‘machine learning algorithm’, we are now able to analyse live order data – predicting a customer’s satisfaction at the very moment of receiving their pizza with over 84% accuracy.
The next phase is to use this data to inform personalised communications to customers in real-time, at the very moment of dissatisfaction. It is truly transformational to react at the moment of feedback, and the prospect is an exciting one for the brand and any businesses that aren’t face-to-face with the customer. We also calculated what an improved customer experience would mean to Domino’s bottom line, revealing a huge incremental uplift from improving feedback scores across each category.