Does Web-based self service reduce telephone calls to call center: an Empirical Analysis

We conduct a field study on a prominent US health insurance firm to examine the cost saving rationale for deploying self service at the call centers. In our setting, firm introduced web self-service portal for its members to contain large telephone handling cost at its call center. Management Practitioners claim that per interaction cost of web based self service channel is an order of magnitude cheaper than the assisted channels of customer service like telephony. However, they do not account for the impact of web based self service usage on the demand for other assisted channels. Surprisingly, the net effect of web based self service on other assisted channels of customer service has not been empirically validated so far. Drawing from the vertical differentiation literature in economics and human information acquisition and processing theories from psychology, we present a theoretical framework to show that both possibilities may exist (1) the web based self-service substitutes the assisted channels and (2) web based self-service leads to increase in demand for assisted channel. In our field setting, we show through a robust diff-in-diff specification that web based self service usage leads to 23% increase in telephony usage. We also use instrumental variables to account for potential selection of customers into the self service usage. Our results inform practicing managers on many counts (1) although cheaper on per interaction level, web based self-service may not be necessarily cheaper overall (2) Need for integrated management in multichannel customer service setup e.g. web based self service should have an option for escalating the query to other assisted channels (3) Self services should not be seen only as a mean for reducing operational cost, as it provides wider options to customers that may lead to higher customer satisfaction and loyalty.