Do firms or the customers always benefit when an electronic retailer increases its IT capacity or advertising effectiveness? Professor Subodha Kumar writes from his research.
Over the last decade, the number of Internet users has increased by leaps and bounds reaching 1.97 billion worldwide in June 2010 – up from only 45 million in 1995 and 361 million in 2000. Following this rapid growth in Internet usage, the number of online shoppers has also risen dramatically. For example, more than 627 million people worldwide have shopped online by October 2005. Furthermore, the US retail e-commerce sales grew 11% in 2009 to reach $155.2 billion and are expected to reach $248.7 billion by 2014.
Commensurate with the increasing number of online shoppers, there has been an explosive growth in the number of electronic retailers worldwide in last few years. An aspect of online shopping that is of interest here is the presence of processing delays at e-commerce sites. There are several potential causes of the delays associated with serving customers at an e-commerce site. Of these delays, there are some that are outside the control of the firm, e.g., the delay at the client end caused by a slow processor or network connection. The delays that can potentially be influenced by the firm’s decisions could be related to network delays caused by network devices, such as switches and routers, and the network connecting to the server, or server delays caused by delays at the firm’s website. In this research, IT capacity refers to capacity that can be increased by the firm to reduce delays experienced by online shoppers.
In this research, IT capacity refers to capacity that can be increased by the firm to reduce delays experienced by online shoppers.
There is considerable empirical evidence that processing delays indeed matter: online shoppers have been found to be very sensitive to processing delays at websites and are prone to abandon shopping if the processing speed is slow. Websites with lower download delay are typically accompanied with greater perceived success by site users. Around 69.4% of all potential online transactions are abandoned and one of the biggest culprits is the poor response time associated with satisfying a customer request. Zona Research Study estimates losses associated with response times of eight seconds and higher to be $4.35 billion annually, whereas the average response time for top 15 e-retailers in July 2005 was 20.16 seconds. A survey shows that the consumers shopping via a broadband connection are even more impatient and will wait no more than four seconds for a web page to render, and 33% of dissatisfied online shoppers attribute their dissatisfaction to the website being too slow. A recent study finds that quoting the lowest price increased the overall satisfaction in only 5% of the top 100 online retail sites, but the site experience, especially performance, provided the biggest payback to retailers.
The examples mentioned above seek to persuade the reader that Information Technology (IT) capacity could be a significant bottleneck in many e-commerce sites – sometimes the traffic arriving at the site may be too high for the installed level of IT capacity. Thus, the goals of reducing delay and generating more traffic are usually in conflict with one another. While more traffic can be generated through advertisement, the higher traffic can slow down the site so that the conversion of arrivals to purchases may suffer. Given that the IT capacity limitations can adversely affect the revenue of e-commerce firms, electronic retailers must factor this constraint while choosing the optimal level of advertising. Most previous studies, however, have ignored the interaction between the IT capacity of a firm and the corresponding optimal advertising level. As IT matures and becomes an integral part of firm’s operations, its role in traditional business decisions needs to be re-evaluated. This role is especially relevant for an online business when deciding on the optimal advertising expenditure for the firm. In this study, we analyse the advertising effort for electronic retailers in a duopolistic setting where the two firms have different IT capacities.
A recent study finds that quoting the lowest price increased the overall satisfaction in only 5% of the top 100 online retail sites, but the site experience, especially performance, provided the biggest payback to retailers.
In a typical e-commerce website, similar to a processor sharing system, the IT capacity constraint affects the processing time of requests submitted by customers. This limited capacity could potentially lead to customer reneging and loss of revenue. To estimate this loss, we consider three possible outcomes of a customer’s visit. First, the customer may make a purchase. Second, the customer may browse the site, but at the end, decide not to make a purchase because of reasons other than a slow response time. Finally, the customer may leave the website before the purchase decision because of slow response time. In this study, we focus on the third scenario or customer reneging.
Several studies have shown that reducing the number of customer requests (for processing) has the single biggest impact on improving response time and reducing reneging. If reneging is excessive, it may be better to reduce the level of advertising and divert some resources to improve response time. We show that the traditional advertising decisions dramatically change in the presence of IT constraints. We consider IT capacity constraints at both firms and obtain optimal advertising paths (and the corresponding traffic) for the firms. Our analyses clearly demonstrate that, in the presence of IT capacity constraints, advertising strategies must be carefully managed by electronic retailers.
Controlling Wasteful Expenditures
Our study provides several interesting results for the managers of the electronic retailers. First, we show that, under certain conditions, as the IT capacity of a firm increases, the total number of customers finishing the transaction at both of the firms decreases. In other words, increase in the IT capacity of a firm may alienate more customers. This is a counterintuitive and surprising result. However, it can be explained as follows. Whenever a firm increases its IT capacity, it advertises more in order to attract more customers from the other firm. Under certain conditions, the firm may not be able to handle the additional traffic in an efficient manner. Hence, in those conditions, the total number of customers finishing the transaction may decrease. We further analyse the conditions under which such scenario is possible. We find that it is possible only when the customers are highly impatient and the IT capacity of the other firm is high.
The main takeaway from the above discussion is that without IT capacity constraints, there is no negative impact on consumers; each firm tries to attract more traffic, but at the end, each consumer is served by either of the two firms. However, in the presence of IT capacity constraints, the equilibrium levels of advertising could not only hurt the profits of the firms, but more IT capacity could increase the dissatisfaction among the customers. In this scenario, a policy-maker may want to intervene by introducing some mechanisms that ease wasteful advertising expenditures (e.g., by imposing a Pigovian tax – A special tax that is often levied on companies with negative externalities.)
Next, we find that, an increase in the advertising effectiveness of a firm may also reduce the number of customers finishing the transaction. The condition under which it happens is same as that in the case of IT capacity, i.e., when the customers are highly impatient and the IT capacity of the other firm is high. Furthermore, since the marginal benefit of advertising increases with an increase in the advertising effectiveness, we would expect advertising to increase with the increase in advertising effectiveness. However, when the marginal cost of advertising exceeds the marginal benefit, the firm will reduce its advertising level. Because of these two opposing effects, we find that the advertising expenditure rate does not always increase with advertising effectiveness. Specifically, the advertising rate increases with the advertising effectiveness at the lower level of advertising effectiveness, whereas it decreases at the higher level of advertising effectiveness. This result can be explained in the following economic terms. When the effectiveness of a resource initially increases, more of it gets used because its marginal benefit continues to be higher than the cost, but as further use of it is made, diminishing returns set in and its use begins to decrease. In summary, the firms need to be prudent in spending dollars on advertising as well as on improving advertising effectiveness.
We also found another interesting and surprising result: As the average value obtained by the firms from each customer increases, the advertising expenditures of both firms increase but their arrival rates remain unchanged. This result may seem counterintuitive at first glance but can be explained as follows. As the average value increases, it encourages both firms to advertise aggressively in an attempt to attract more customers. However, while the arrival rates (Average rate at which customers arrive to the system) for either firm do not change, their advertising costs increase. In summary, an increase in average value triggers unhealthy advertising competition between the firms. Again, this result alludes to the fact that the firms need to control wasteful expenditures under competition.
Customers’ Impatience Level
We find that that the higher capacity firm gains when customers are more impatient. Based on our results, we conclude that as the impatience level increases, the increase in reneging for the higher-capacity firm is more than the increase in its arrival rate. Unless the possibility of market expansion is considered, in traditional advertising problems, the loss of one firm is always the gain of the other firm, i.e., the total loss in the system is zero. On the other hand, our results show there is a positive loss in the system in presence of the IT capacity constraint, because customers can renege. This is an important result for managers because it highlights the significance of considering IT capacity constraints and customer behavior at e-commerce sites in what are traditionally considered to be pure marketing decisions.
Further, when customers become more impatient, both firms reduce their advertising rates. Since reneging increases substantially with impatience and one firm does not gain everything from the loss of the other firm, it is not advantageous for either firm to advertise more with an increase in the impatience level. We also note that the difference between the advertising expenditure rates of firms increases at higher impatience levels. This result indicates that when impatience levels are relatively high, the more effective and the higher capacity firm should advertise more as compared to the less effective and the lower capacity firm.
Finally, we analyse the scenario when the firms are very similar in terms of IT capacity and advertising effectiveness. In this case, the marginal benefit of increasing advertising (following an increase in the advertising effectiveness) is always more than its marginal cost. This result is clearly different from the results discussed earlier for the case when the firms are not similar. Earlier we also discussed that, under certain conditions, increasing IT capacity of a firm may result in higher customer dissatisfaction. However, when the firms are similar, increasing IT capacity always makes the customers happier and more people complete the transaction.
For a more detailed analysis, see the following article: D. Liu, S. Kumar, and V.S. Mookerjee, “Advertising Strategies in Electronic
Retailing: A Differential Games Approach,” Information Systems Research, forthcoming. This article contains edited and summarised excerpts from the original article.