Predictive models for targeting
Our predictive modeling approach enables precise, targeted marketing activations. We routinely score segments within a client’s CRM database or externally in media panels, via look-alike models or even at the household level across all U.S. households. In each scenario, the goal is to enable 1-to-1 targeted marketing, custom experiences and unique brand activations.
We first start by defining segments in the context of the category. Once the segments are defined, we build a representative data set and then use in-depth analytics to uncover the most useful segment predictors from various input sources.
The modeling outputs allow us to classify individual customer records or households into segments, thereby enabling 1-to-1 targeted marketing across activation methods.
Client benefits: Our clients typically start by using our predictive models to enable 1-to-1 communication within their CRM database. Scoring their database with customer segments enables tailored digital activations, emails, direct mail, and even fully customized website experiences based on a customer’s category related motivations and interests. Clients then leverage our total population models to target potential customers via precise campaigns and activations that can be 1-to-1 at the household level, or by using more geo-targeted activations based on block-groups, metro areas and store clusters.