This article presents the evolution of logistics performance in the B2C e-Commerce sector in Brazil throughout the year 2000. For this purpose, two surveys were carried out to survey the following aspects related to customer service: cycle time, deadline consistency delivery times, the number of failures and the use of logistic operators to deliver the products. The analysis was based on the hypothesis of a trend towards improvement in logistics performance over time by retailers due to the adoption of more advanced technologies in response to the logistical problems that occurred in the country during the period studied. The results showed that retailers in general improved their performance in terms of cycle time. In terms of damage, there was no significant variation in the period studied. On the other hand, the delivery time consistency attribute worsened significantly due to the increase in the gap between the promised time and the actual delivery time. In addition, the use of Correios for deliveries has been completely replaced by the use of specialized logistics operators.
- INTRODUCTION
Many controversies arise when talking about the future of e-Commerce in Brazil. Very few virtual retail companies have shown positive results; most still dream of generating profits and reaching investment break-even. However, there seems to be a common point that everyone agrees on: the order fulfillment process and distribution logistics are the main bottlenecks in e-commerce. We are often faced with news in the press with reports of negative experiences lived by consumers in purchases made over the Internet: delays in relation to the promised delivery date, products delivered with faults, billing errors and cancellation of orders due to lack of products are just some of the experienced problems that can generate negative experiences in virtual shopping.
Virtual retail logistics presents some singularities that are not present in traditional retail, such as, for example, the integration between information about the availability of a certain product through the website – front-end – and the actual availability of this product in stock – back-office (Reynolds, 2000). However, despite constant reports of dissatisfaction, there are few studies that seek to identify the main causes of e-commerce to justify this performance. During the year 2000, due to media exposure, logistics was placed as one of the key issues for the success of virtual companies, which led companies to invest in better systems and processes to meet the new needs of virtual retail. Looking for ways to differentiate themselves from their competitors, they discovered that Logistics could be used as a powerful weapon in order to respond to their customers' needs.
According to Peter Drucker (1999), electronic commerce is to the Information Revolution what the railroad was to the Industrial Revolution and is characterized by the electronic exchange of information. Therefore, B2C (business-to-consumer) e-Commerce is any form of electronic commerce involving a business and the final consumer. This type of commerce is increasingly a powerful retail channel. According to a survey carried out by the IDC research institute (2001), B2C e-commerce should move around US$ 470 million this year in Latin America, reaching around US$ 1,6 billion in 2003.
 |
- CONCEPTUAL BASES
Two basic concepts were used to guide the research: customer service and order cycle. According to the concept of customer service, perceived service quality results from the comparison between customer expectations and the performance of the service provider, based on a set of dimensions defined from studies by La Londe et al (1988), Bowersox and Cooper (1992) and Christopher (1992): product availability, order cycle time, lead time consistency, delivery frequency, delivery system flexibility, fault recovery system, back-up information system, support in physical delivery and post-delivery support. In this work, only the main results obtained given by the following dimensions will be considered: the order cycle time, the consistency of the delivery time, the breakdowns and the physical delivery company.
The supplier's performance is the result of a set of activities developed during the order cycle, that is, the process that begins when the customer begins to prepare his order, and ends when, finally, takes possession of the purchased goods and considers the transaction satisfactorily completed. The order cycle involves four activities: order preparation and transmission, order processing, order selection and transportation, and delivery to the customer (Bowersox and Closs, 1996). Therefore, the total order cycle and its variability is the result of cycles of activities involving information management and physical flow management.
This work is based on the results of two surveys carried out at the beginning and at the end of 2000 with the objective of knowing the logistical performance of virtual stores in a series of customer service attributes. The purpose of this document, therefore, is to study the evolution of the performance of virtual retailers of consumer goods and contribute to the improvement of e-commerce logistics in Brazil, through the identification of the main problems and their causes.
 |
- METHODOLOGY
The first survey, conducted during the months of April and May 2000, was carried out with three of the best known virtual retail companies in Brazil. Orders were made for CD's of Popular Brazilian Music (MPB), through the websites of the chosen companies, indicating for the place of delivery different addresses of cities located in all geographic regions of Brazil, that is, North, Northeast, South, Southeast and Center -West, including capitals and inland cities. In all, 30 volunteers were selected, each one residing in a different city, who volunteered to receive 3 CD's in their homes – one from each evaluated company – and answer about dates and conditions for delivering the products. According to the original sample, a total of 90 CDs would be purchased. The expected sample, however, suffered some reductions due to the withdrawal of research participants and problems in carrying out the financial transaction. In the latter case, there were 3 orders that were not carried out due to problems in releasing credit and 3 orders whose slips were not paid because it was not possible to identify, in a timely manner, the codes of the orders to which they referred. The final sample now has 71 orders paid for and monitored through status checks and contacts with volunteers at delivery addresses.
 |
The second survey, carried out in December 2000, had the inclusion of another virtual store and 19 delivery addresses in the city of Rio de Janeiro were selected, and the CD's chosen were MPB and National and International Pop/Rock. Each of these participants volunteered to receive 4 CD's – one from each evaluated company – and answer about the dates and conditions of delivery of the products. The initial objective was to carry out 144 orders; however, due to sample losses, only 131 orders were actually completed and paid, that is, a sample loss of 13 orders. Of these, 5 were unable to complete the transaction due to the impossibility of printing the bank slip, due to problems with the virtual store's website and due to card credit problems. Another 5 orders had the limit for placing orders exceeded, whose bank slips were not paid or the determined limit date of the purchase exceeded. In addition, 3 orders had data filled in incorrectly by the buyer.
 |
In both surveys, two forms of payment were used: credit card and bank slip. In the second survey, the new virtual store did not have a bank slip payment method, hence the higher number of orders made by credit card. Alongside the information provided by the volunteers, continuous monitoring of each order was carried out, using information obtained from the websites with the aim of examining the performance of suppliers in the various stages of the order cycle, in light of the dimensions of customer service . The objective was to test the Internet shopping system from the consumer's perspective, so that the shopping experience could be portrayed from the point of view of those who are buying. It is good to remember that dissatisfied customers always have the alternative of looking for another virtual supplier, or going back to buying in traditional retail, which is their main point of reference, and with which they will compare their shopping experience in the virtual world.
 |
- PRESENTATION AND DISCUSSION OF THE RESULTSThe results will be presented and discussed considering each of the selected dimensions.
4.1. Cycle time
This dimension measures the number of days that elapses from the moment the buyer places the order until the moment when the requested goods are delivered, that is, from the stage of preparation and transmission of the order until delivery. Cycle time is the picture of the operational efficiency of the entire system of a virtual store. It is worth remembering that the cycle time presented here does not take into account the payment time of the bank slip, which is the responsibility of the buyer.
 |
In general, the delivery performance of Internet sales sites improved by around 42% from May 2000 to December of the same year, although it was possible to expect that in December the performance could be worse due to the increase in sales at Christmas.
Analyzing the evolution of delivery time by means of payment, it is observed that companies are more efficient both for purchases through bank slips and for purchases by credit card. In the case of bank slips, there was a reduction of around 43% and a reduction of 33% for credit cards.
 |
As seen, one of the components of the total order cycle is the order processing activity. A significant change in the ordering processes of the virtual stores was the adoption of the bank slip printing alternative. In the survey carried out in May, one of the stores did not have this option. The bank slip was delivered to the billing address, contributing significantly to increasing the total cycle time by approximately 5,3 days. In the case of orders made by credit card, it was possible to identify the time required for the credit to be confirmed. An average reduction of 1,2 days was observed, contributing to the reduction of the total delivery time.
 |
4.2. Delivery Time Consistency
This dimension is related to the uniformity of performance in meeting delivery deadlines. The indicator used was the percentage of late deliveries. Graph 1 shows the results of the evolution in performance in terms of meeting these deadlines.
 |
The delay is a measure of the adequacy of the promised delivery time with the actual delivery time. When we calculate how many deliveries were made after the promised deadline, we see a significant increase in the percentage of delays from May to December. As mentioned earlier, the total delivery time has been reduced. However, the adequacy of the companies' promise has deteriorated. The inadequacy of promises can cause frustration of customer expectations and can be more harmful to the virtual company than having a higher delivery time. This increase in the gap between the promised deadline and the total realized cycle time can be seen in Graph 2 below.
 |
It is clearly seen that the reduction in promised delivery times was not accompanied by a reduction in actual delivery times. Thus, although delivery performance has improved from May to December, the promise of reducing this time by companies was not achieved.
4.3. breakdowns
A fault is identified when the product received by the buyer is not in perfect condition. From April/May 2000 to the second survey in December, no significant variation was observed in the percentage of breakdowns.
 |
Due to the increase in Christmas sales volume, it was expected that there would be a significant increase in the size of breakdowns, which did not occur. One of the factors that can explain this stability in the percentage of failures is the change in the company that delivers orders placed over the Internet, as seen in the following dimension.
4.4. Physical Delivery
One of the important aspects of physical delivery was the significant increase in the use of logistics operators specialized in express deliveries to the detriment of the use of the Post Office. According to graph 3, in April/May 2000, Correios had delivered around 76% of orders placed. In December of the same year, graph 4, the use of Correios was not observed during the survey. Graphs 3 and 4 below show the main delivery companies in April/May and December.
 |
Two factors can explain this fact: the preoccupation of virtual stores with meeting the deadlines promised for Christmas, hence the use of specialized logistic operators, and the fact that the December survey was carried out only in the city of Rio de Janeiro, where performance of logistics operators is strong. According to an executive from one of the delivery companies, the logistics operators intend to serve fillet mignon first, which are the country's big cities, since Correios has great capillarity and experience in deliveries in the most distant areas of the country's metropolises.
- CONCLUSION
- The April/May 2000 survey was carried out nationwide. Due to the sample size and the number of requests for each city, it is not possible to state that the difference between cities is significant. Therefore, one of the limitations of the study is the fact that the April/May 2000 survey was carried out for the whole country and the December 2000 survey only in the city of Rio de Janeiro.
- Disregarding this factor, we can say that there was a significant improvement in cycle times, that is, in the total delivery time. This evolution is present for the two forms of payment analyzed: bank slip and credit card.
- Despite the improvement in cycle time, there was a significant reduction in consistency in the operational performance of the virtual stores. The reduction in the promised delivery time was not accompanied by a significant reduction in the total cycle time. As a consequence, there was an increase in delays from around 4% to 48%. This inadequacy of the promise of virtual stores, in a customer-oriented view, is extremely harmful to the level of service provided. It is often preferable to promise a delivery in a longer time than to delay a delivery in a shorter time, since the expectation of service is frustrated. This frustration can be even more aggravated when this delay is a Christmas gift, when the cost of non-delivery is more linked to the emotional aspect of the consumer.
- One of the factors that reduced the order cycle was the significant reduction in order processing time. In the case of bank slips, one of the virtual stores now offers the option to print the slips, thus eliminating the activity of delivering the slips to the billing address, which took an average of 5,3 days. In this way, changing the processes of one of the virtual stores contributes to reducing the total order delivery time. Furthermore, in the case of credit cards, there was a significant increase in operational efficiency in the credit confirmation stage. This led to an average reduction of the total credit card purchase cycle by 1,2 days.
- An increase in the percentage of breakdowns was also expected due to the increase in sales volume during the Christmas season. However, this variation was not significant.
- One of the factors that may have contributed to this stability in the percentage of breakdowns is the significant increase in the use of specialized delivery companies, contributing to safer physical movement than that of the Post Office. This accounted for around 76% of deliveries made in April and May. In December, Correios was completely replaced by specialized delivery logistics operators.
There is no doubt that e-commerce is facilitating access to products, leading to more dispersed sales and increased logistical complexity. However, care must be taken not to take a narrow view of the order cycle. Transport times are not the only determinants of total delivery time. The main causes of increased delivery times are found in the order processing and fulfillment phases. The speeding up of credit confirmation time was an example of evolution in the operational performance of virtual stores. Having a good performance in the steps that make up the order cycle can generate 3 results: increased sales, a result of less dropouts when taking the order and a reduction in unavailability, cost reduction, due to the reduction of product returns (logistics reverse), and improved service levels, achieved by reducing cycle time and providing accurate and fast information. Therefore, companies must learn to deal with the trade-off between having a variety of options and controlling the flow of information.
BIBLIOGRAPHY
Bowersox, DJ; Closs, DJ Logistical Management – The Integrated Supply Chain Process. New York, NY: McGraw-Hill, 1996.
Bowersox, DJ; Cooper, MB Strategic marketing channel management. New York, NY: McGraw-Hill, 1992.
Christopher, M. Logistics and Supply Chain Management. London: Pitman, 1992.
Drucker, P. – Beyond the Information Revolution
Reynolds, Janice – Logistics and Fulfillment for e-Business: CMP Books, 2000
LaLonde, BJ; Cooper, MC; Noordewier, TG Customer service: a management perspective. Oak Brook, IL: Council of Logistics Management, 1988.
