Predictive modeling uses known data results to develop models that can predict future values that inform business decisions.
Given known variables of a complex set of marketing transactions, a predictive model can supply missing variables based on past performance.
For example, given a marketing and financial return data set for multiple channels, the individual outcomes of each can be compared against one another to determine the best tactics for driving revenue. Using marketing initiative data (CPC, site visits, add to cart conversion, checkout conversion) and associated financial data (initial ad cost, transaction totals, margin, and LTV), the model can determine total investment return across multiple channels. Consequently, brands can select highest performing initiatives in which to invest.
Learn more about predictive analytics here.