MKTATTRIBUTION Procedure

The MKTATTRIBUTION procedure supports different types of market attribution models, which have been used to identify which marketing channels drive customer conversions and to help optimize the investment in those channels. Customers are "converted" when they are persuaded by marketing content to take a particular action, such as buying a product, that a business wants them to take. As the array of media channels where businesses can promote their products and services grows, most customers engage with marketing content through multiple channels, such as in television advertising and in ads on social media. Businesses want to know to what degree each channel contributes to their marketing success. This is called the multichannel attribution problem. The Markov attribution model (MAM), one of the market attribution models, approaches the attribution problem in a probabilistic way by using a Markov chain, a particular stochastic process in which the probability distribution of any next state depends only on what the current state is, regardless of any preceding states. The MAM uses a first-order Markov chain to calculate the probability of interaction between pairs of channels in the customer journey and to evaluate the channel’s contribution to customer conversions through the so-called removal effect. The removal effect would enable a business to give a reliable assessment of the marketing contribution of each channel. In addition, the MKTATTRIBUTION procedure provides a number of heuristic attribution models: the first-touch, last-touch, linear, position-based, and time-decay attribution models.

PROC MKTATTRIBUTION requires SAS Cloud Analytic Services (CAS) in order to run, and it does the following: enables you to run on a cluster of machines that distribute the data and the computations, and exploits all the available cores and concurrent threads.