Eva Ascarza
WORKING PAPERS / UNDER REVIEW
WORK IN PROGRESS
SELECTED ABSTRACTS
The ability to retain existing customers is a major concern for most businesses. However retention is not the only quantity of interest; there are other behaviors that influence the value of a customer. In most contractual situations, the exact revenue that will be generated by each customer is uncertain at the beginning of the contract period; rather, customer revenue is determined by how much of the associated products or services each individual uses. While a number of researchers have explored the problem of modeling churn in a contractual setting, there is surprisingly limited research on the modeling of usage while under contract.[go to the top]
The ability to retain existing customers is a major concern for most businesses. However retention is not the only quantity of interest; there are other behaviors that influence the value of a customer. In most contractual situations, the exact revenue that will be generated by each customer is uncertain at the beginning of the contract period; rather, customer revenue is determined by how much of the associated products or services each individual uses. While a number of researchers have explored the problem of modeling churn in a contractual setting, there is surprisingly limited research on the modeling of usage while under contract.
The ability to retain existing customers is a major concern for most businesses. However retention is not the only quantity of interest; there are other behaviors that influence the value of a customer. In most contractual situations, the exact revenue that will be generated by each customer is uncertain at the beginning of the contract period; rather, customer revenue is determined by how much of the associated products or services each individual uses. While a number of researchers have explored the problem of modeling churn in a contractual setting, there is surprisingly limited research on the modeling of usage while under contract.
In many service industries, firms introduce three-part tariffs to replace or complement existing two-part tariffs. As opposed to two-part tariffs, three-part tariffs offer allowances, or “free” units of the service. Behavioral research suggests that attributes of a pricing plan may affect behavior beyond their direct cost implications and there is evidence that customers value “free” units above and beyond what would be expected based on the change to their budget constraint. Nonlinear pricing research, however, has abstracted from such an effect.[go to the top]
Using a unique data set, we analyze tariff choice and usage behavior for customers who switch from two-part to three-part tariffs. We find that switchers significantly “over-use” compared to their prior two-part tariff usage. They attain a level of consumption that cannot be explained by a shift in the budget constraint. We estimate a discrete-continuous model of tariff choice and usage that accounts for the valuation of “free” units. Our results show that 81.9% of three-part tariff users have a positive valuation for the new attribute. This translates into a 12.2% revenue increase.
The customer migration model is a simple model to forecast customer behavior in a noncontractual setting. As such, it has been proposed to compute Customer Lifetime Value under these circumstances. This model can be expressed as a Markov chain, where the probability of future transactions depend on past customer behavior, typically summarized in terms of recency, frequency, and monetary-value characteristics. Even though the model has been applied in several business situations, its forecasting accuracy has not been documented. In this paper, we undertake such a validation exercise and find that the customer migration model gives accurate predictions only at the aggregate level. If, however, this model is used to predict individual-level behavior, the forecasts are generally poor, leading to suboptimal decisions if relying on this model at the strategic level.[go to the top]
In many service markets, firms offer three-part tariff contracts in conjunction with, or as a replacement to, existing two-part alternatives. Understanding customer switching behavior between these pricing structures is crucial to drawing conclusions about firm profitability. However, even though the literature on the adoption of multi-part tariffs is extensive, to date there has been no empirical research that has examined actual switching behavior from two-part to three-part tariff plans.[go to the top]
In this research, we first analyze a unique dataset containing 12 months of usage and tariff choice behaviour for a sample of mobile phone subscribers from a telecommunications company in Asia. The uniqueness of this data lies in the fact that we observe a natural experiment in which, after having observed consumers choices under two-part tariff plans, the company launched a three-part tariff scheme, while retaining the existing two-part tariffs. This market environment allows us to identify which factors ultimately determine customer switching behavior between these two pricing schemes. Consistent with economic theory, we find that light users are more likely to switch to three-part tariffs, which in itself justifies the inclusion of this type of contract. We also observe behavior that is consistent with the various “flat-fee” biases previously documented in the literature. To confirm (or discard) the underlying psychological mechanisms, we complement this analysis with a series of focused lab experiments.