A financial institution wishes to optimize the mix of new customer acquisition channels, limiting the usage of higher-cost direct-mailer channels to just those customer segments that have low conversion propensity when targeted via other relatively lower-cost channels. Segmentation techniques and predictive modelling are used to identify customer characteristics that drive high conversion propensity for direct-mailer campaigns and low propensity for other channels. The resulting customer acquisition model enables the financial institution to engage in effective prospect selection for its acquisition campaigns across the various channels. This allows the company to enhance the effectiveness of its direct-mailer campaigns as well as increase the marketing RoI for acquisition campaigns as a whole.
A financial institution targets new customers through acquisition campaigns conducted via several online & offline channels, such as direct mailers, electronic mailers, SMSs, etc. The company uses a mix of high-cost and low-cost channels to control customer acquisition costs. Channels like direct mailers, while effective as new customer acquisition tools, entail higher costs. Electronic mailers and SMSs are relatively more cost-effective.
In the financial institution’s experience, different customer segments respond differently to the various acquisition channels. For instance, some customers respond positively to direct mailers but not to campaigns via other channels. The same holds true for the electronic mailers and SMS channels. In addition, there exist customer segments whose response to the company’s acquisition campaigns is fairly uniform across the several communication channels.
In an effort to control new customer acquisition costs, the financial institution wishes to optimize the channel mix for its acquisition campaigns. Specifically, the company wants to limit the usage of relatively higher-cost channels to only those customer segments that are likely to respond positively to campaigns via such channels and whose conversion propensity reduces markedly when other relatively lower-cost channels are used.
The propensity of potential customers to respond positively to acquisition campaigns via different channels depends on customer demographics (age, gender, academic background, professional profile, income level, marital status, family size, etc) as well as the type of financial product being pitched and the acquisition channel being used for the pitch.
Exploratory analysis is conducted on the various customer acquisition campaigns run by the financial institution. Historical data is examined for campaigns across product types, acquisition channels, cities in which the campaigns were run, and a wide range of customer profiles that were targeted. Segmentation techniques are deployed to understand the underlying factors that determine the variation in customers’ propensity to respond positively to campaigns over the different acquisition channels.
Predictive analytics techniques are used to develop customer acquisition models that identify prospects who are likely to get converted only when targeted exclusively via direct mailers, and not otherwise. Such predictive modelling enables the financial institution to decipher customer characteristics that drive high conversion propensity to direct-mailer campaigns and low conversion propensity to other acquisition channels.
With such customer acquisition models in place, the financial institution is able to put in place criteria to select appropriate acquisition campaign channels that optimize acquisition costs as well as enhance the conversion propensity among the prospects. Better prospect selection for the various campaign channels helps the company maximize the impact of its higher-cost direct-mailer campaigns as well as improve the marketing RoI for its customer acquisition campaigns as a whole.