Find twins for your best customers through advanced statistical calculations
The starting point for a scoring analysis is to define the customer group to which you want to find twins (the target group). Through your historical customer data regarding, for example, purchase date, number of purchases, purchase price or recruitment channel, a target group is defined. We test and evaluate a number of algorithms and build the scoring model that is optimal based on your unique customers. The model then ranks prospects in a defined target market based on the probability that they are similar to those included in the customer group.
The strength of a scoring model as a selection tool is that it uses all our available data in the search for the prospects who best match your customers.
Predictive model in several dimensions
This is the most advanced selection method. Instead of just looking at a target group (for example your best customers), we combine several desirable characteristics. It can be a combination of, for example, responsive customers, customers who are loyal and customers who are good payers. It can be seen as building specific models for each property we want to capture. The models are combined into a final model that works in several dimensions.