Empirical study on cost components of E-com product returns in footwear industry

(1)

1

Empirical study on cost components of E-com

product returns in footwear industry

Xingni Chen

S2970139

x.chen.26@student.rug.nl

Groningen

Jan 2018

MSc Technology & Operations Management

Faculty of Economics and Business

University of Groningen

First Supervisor: Dr. K.J. Roodbergen

Second Supervisor: Dr. N.D. van Foreest

(2)

2

Abstract

Purpose – The goal of this paper is to identify costs of E-com product returns in the footwear industry. And then based on the identified cost components, adding two most common factors in E-com, return policy and free shipping promotion. By determining how return policy and free shipping promotion affect the sale revenue and cost of returns to develop retailer an expected profit model.

Design/method/approach- The paper follows the three phases in the research study; 1) performing standard case study to identify all factors that impact costs of returns, using literature as input to build a coding tree; 2) using ABC costing method to organize the cost factors and quantify them 3) studying crocs and filling out the framework with actual cost number to verify applicability.

Findings – Cost components related to the reverse logistic, DC process and customer service are identified according to the product return flow. Furthermore, the trade-off between sales revenue and cost of returns are quantified in the model after considering the return policy and free shipping promotion.

Originality/value – This is the first empirical study to investigate the topic of cost of product returns, and extend to investigate the financial consequence affected by return policy and free shipping promotion. Regarding the practical contribution, the model can assist manager make the managerial decision on return policy and free shipping promotion to maximize the profits.

Keywords: product returns, costs, reverse logistic, lenient return policy, return deadline, free shipping promotion, threshold

(3)

3

Acknowledgments

This thesis is my final project for the master studies in Technology & Operations management at the University of Groningen. Several people have contributed to writing this thesis and deserved to be acknowledged and thanked.

Foremost, I would like to thank and express my gratitude to my first supervisor Kees Jan Roodbergen., Thanks for his patient to coaching me, guiding me and teaching me a lot throughout the whole period for my thesis writing. I also appreciate all frequent meetings with him; those meetings make me stay motivated and also direct me the way when I was in a dilemma. Thanks to his critical and constructive comments, push me to work hard and improve my thesis. Also thanks to my second accessor, Nicky van Foreest, thanks for his comments, which challenge me and give me different perspectives to work on my thesis.

I really appreciate Crocs and my company supervisor Sascha de Bruijn. Thanks for giving me an internship opportunity, and allowed me to extend my internship to do my graduation project in the company. Thanks for their support, and data sharing. Thanks to Sascha, giving a detailed explanation of the company structure and direct me to the right person for the right topic, he is always there when I have any doubts about processing my empirical study.

Last but not least, I am especially grateful to all interviewers who give me all support and input for my thesis. Without their participation, this thesis would not have been possible in a way it is presented.

Jan 2018, Den Haag

Xingni Chen

(6)

6

List of Abbreviations

ABC: Activity-based cost

A/B test: Sometimes called split testing, is comparing two versions of a web page by showing the two variants to similar visitor at the same time

AOV: Average Oder Value

ASP: Average Selling Price

CFS: Contingent Free Shipping

CR: Case Reserve

CS: Customer Service

DC: Distribution Center

EOL: End of line

KPI: Key Performance Indicator

MAC: Moving Average Cost

RMA: Return Merchandise Authorization

UFS: Unconditional Free Shipping

WH: Warehouse

(7)

7

List of tables

Table 1: List of the department and personnel interviewed in Step I ... 18

Table 2: Overview of Coding Tree ... 19

Table 3: The documents read as part of research at Crocs for Phase 3... 21

Table 4:Cost components for cost framework ... 23

Table 5: Interview Summary about effects of time leniency and free shipping ... 24

Table 6: Setting of cost components ... 25

Table 7: Framework about the revenue ... 27

Table 8: Framework about returns ... 28

Table 9: Summary about cost components on product returns ... 32

Table 10: Summary of product returns costs... 33

(8)

8

List of figures

Figure 1: Crocs product return process ___________________________________________________________________ 30 Figure 2: Crocs DC process about product returns _______________________________________________________ 31 Figure 3: Warehouse layout in Crocs ______________________________________________________________________ 31

1. Introduction

(9)

9

industry is quoted between 20% and 35% in 2014, and this number is still increasing (Van Coolwijk, 2014). Many substantial costs like reverse logistical cost, product depreciation, management of return process and so on (Blanchard, 2005) are hidden behind each return. Nick Robertson CEO of ASOS, a leading online fashion company mentioned in 2012 that, a 1% decrease in return rates could immediately lead to a 10 million pounds profit, which accounts for approximately 30% of their net profit in 2012 (Thomason, 2013). It appears product returns may signal decrease in profits, and it is time to investigate how much does it cost for product returns.

From the previous research, product returns are considered as a costly expense for online retailers (Cai, 2012; Charlton, 2007; Petersen, 2009). It starts with the customer, shipping to the warehouse for quality checking, restoring, repackaging, and then it is readied for new owners (Li et al. 2008; Roger 2001; Stock & Shear 2002). A product to be returned can cost double or even triple than that in its forward supply chain (Bonifield et al., 2002; Ganeshan et al., 2004). To assist companies to comprehend their costs spent on reverse logistic, cost components of product returns have to be determined. What’s more, in order to reduce product returns, most companies just create product-return disincentives, such as limited return deadline (say, within 15 days after purchase) and nonrefundable purchase costs (shipping costs or restocking fees) (Su, 2009). Setting restrictions on policies is also suggested in many research, they explain that extended return deadline and free shipping can lead to increasing online product returns (Charlton, 2007; Haarlander, 2001; Davis, Hagerty, & Gerstner, 1998). However, other researchers think different they argue that allowing customers to return products with impunity can reduce customer risk and increase customer satisfaction.This will result in increasing the number of purchase and raise the company’s revenue from sales (Su, 2009; Che 1996; Heiman, 2001). Therefore, the company applies a lenient return policy, and free shipping promotion seems like a trade-off between sales and product returns.

(10)

10

figure out the financial consequences of product returns (Petersen, 2010). This research aims to investigate the cost components for product returns firstly, and then create a framework to calculate the lost value of product returns. On top of that, this research also investigates the changes in financial consequences that arise from the various shipping and return policies. Ultimately, this would help readers to understand how profit changes when online retailer provide different information about return and shipping policy, after considering return costs.

The goal of this research is to create a quantitative framework for identifying costs of E-com product returns in the footwear industry and testing this framework empirically.

To assist the research topic, following sub-questions are framed. - What factors influence the cost of product returns?

- How do these factors influence the cost of product returns?

- How can these factors best be quantified?

- Is the model applicable in a practical setting?

Taking limited literature on this research subject into account, the author would like to conduct a case study in a footwear company, Crocs. It provides an exquisite chance and ideal environment to conduct semi-structured interviews to research what possible factors and how these factors influence the cost of product returns. The research intends to build a cost framework can base on data gathered from the interviews. Furthermore. Crocs also provide the opportunity to apply the real data to test and verify the application of this model.

(11)

11

2. Theoretical background

This chapter aims to give a literature background on the research subject. The chapter starts with a theory of the process of reverse logistic, then extends to different cost method to show a concept how to allocate actual costs in a manufacturing industry. Further, how other indirect factors affect the cost of product returns are discussed in section 2.2.

2.1 Direct Factors to determine cost components

Product returns are considered as a hassle for an e-business’s supply chain management and a drain on overall profitability (Charlton, 2007). To figure out how much costs are hiding behind the product returns, the author expects to contribute to literature about cost components of product returns. However, most literature only illuminates the processes that occur in reverse logistic rather than the cost components of product returns. Hence, this research applies the theory about reverse logistic as the background of direct factors to determine cost components, and then use ABC method to calculate the actual costs incurred in product returns.

Reverse logistic and Product Returns

Contrary to reverse logistic systems, reverse logistic is the flow and management of products, packaging, component, and information from the point of consumption to the point of the supplier (Li & Olorunniwo, 2008; Rogers & Tibben-Lembke, 2001; Stock & Shear, 2002). In accordance with the definition of reverse logistic, it illustrates that a reverse supply chain is much more complicated than forward supply chain since many more activities are included. Loana (2013) indicates that processing returned merchandise includes damaged products, seasonal inventory, restock, salvage, recalls and excess inventory. Some barriers exist in these activities while processing returns. For example, the uncertain timing of returns, the need to balance returns with demands, and the uncertainty in materials recovered from returned items are also some of the complicating characteristics impacting production planning and control for remanufacturing (Fleischmann, 2001). Furthermore, some companies engage with the third party in the network planning in return processing to improve efficiency and reduce handling costs (Guide, 2000; Srivastava, 2008; Koloszyc, 1999). For example, some companies outsource carrier to the third party to reduce company’s pressure on transportation of returned items.

The process which occurs costs in reverse supply chain can be generalized into six key points below (Bowersox, 1991; Chan 2008; Li, 2008)

 Product acquisition – collecting return items from the user  Reverse logistics – transporting the products to a central location  Inspection - Unpacking and assessing the condition of the return

 Disposition – Making the most profitable decision for restorage and reuse these returns  Remanufacturing – Returning the product to original specifications

(12)

12

Costs occurring in each of the processes mentioned above can erode the profit of the company. However, there is a dilemma that product returns are always considered as evil for the E-business. Some recent studies, in fact, began illuminating the potential benefits of numerous return of products. Bower (2006) indicates that customer’s ability to return products may have a positive effect on his or her future purchase and increase long-term profits. In his study, he finds a satisfactory product return experience can provide another touch point for building a successful buyer-seller relationship. Thus, increasing customer satisfaction on product return can increase the number of future purchase and thus raise the company’s profits from sales.

ABC cost

The traditional method of costing relied on the arbitrary addition of a proportion of overhead costs onto direct costs to attain a total product cost. This type of costing system usually allocates costs based on a single volume measure, such as direct labor hours or machine hours, while using this simplistic volume method to allocate overheads as an overall cost driver (Allott, 2004). This can lead to an increase in inaccuracy because it seldom meets the cause-and-effect criteria desired inaccurate cost allocation.

Activity-based cost (ABC) is an alternative approach to the traditional method, which needs to identify appropriate output measure of activities and resource (cost drivers) and their effects on the costs of making a product or providing a service. ABC forces on accumulating costs via activities, then to allocate accurate costs on overheads to products, service, and customers (Barrett, 2005). There are four steps to implement ABC (Friedman, 1995);

1. Identify activities: the organization needs to undertake an in-depth analysis of the operating process of each responsibility center. Each process might consist of one or more activities required to produce an output.

2. Assign resource costs to activities: this involves tracing costs to cost objects to determine where the total cost comes from and why the cost occurred. Cost can be categorized in three ways;

- Direct cost: this cost can be traced directly to a cost objective or one output. For example, labor cost or material cost. It can vary with the rate of the output.

- Indirect cost: this cost that is not directly accountable for an individual output or a cost objective. It can include multiple activities and therefore cannot be assigned to specific cost objectives. Examples of indirect costs are storage costs.

- General/Admiration cost: this cost refers to those are likely to still incurred by company whatever output is produced. For example, salaries of administrative staff, office fixtures, and equipment, or depreciation.

3. Identify outputs: outputs are identified based on a list of activities (bill activity) and corresponding consumer resources. Outputs might be products, services or customers. 4. Assign activity costs to output: this is done using activity drivers. Activity drivers assign

activity costs to outputs according to the consumption or demand for activities.

(13)

13

and activities, like manufacturing industries. Meelah (2017) also presents an automobile manufacturing company as a successful example, which improved their costing accuracy to understand the true costs from projects or other initiatives better after using ABC method. In this research, processing a product return consists of multiple activities, the principles, and philosophies of ABC thinking can provide a more accurate method of costing of products and service. Furthermore, using ABC, organizations can gain a thorough comprehension of overheads and what causes them to occur. This might be beneficial for managers to reduce and eliminate those costly and non-value adding activities after knowing the cost of product returns in each activity. Therefore, ABC cost method is a superior selected approach to identify the costs of product returns in this research.

2.2 Indirect factors influencing cost of product returns

Apart from the costs and prevalence of returns, most retailers offer some benefits hoping that the positive effect on demand will offset the negative effect on returns (Liu & Wei Kwok, 2003). In numerous research, they find return policy and free shipping promotion are two factors can significantly affect the product returns (Griffis et al., 2012; Mukhopadhyay & Setoputro, 2005; O’Neill & Chu, 2001; Van Coolwijk, 2014). Normally, the return policy is a consumer risk reliever often used by retailers, and free shipping is a most common promotion which runs multiple times in a year by retailers. (Petersen, 2009; Haarlander, 2001) .These two factors present a challenge to managers on how to deal with customer product returns to maximize profits (Zhou, 2016). Therefore, next two sections illustrate the effect of lenient return policies and free shipping promotion on product purchase and subsequent returns.

Lenient Return Policy

Based on prior literature, the typology of return policy factors can be identified being lenient along five different dimensions;

1) Time leniency: Normally e-retailers offer the length of time to the customer to return the product ranging from 7 to 90 days, or even longer. Return policies that provide a longer length of time to return products are regarded as more lenient (Davis et al., 1998; Zhou & Hinz, 2016).

2) Monetary leniency: Lenient return policies allow for a refund of the full monetary amount paid for the product, while strict policies allow for only a portion of the purchase price to be refunded, usually by imposing a “restocking fee” or a non-refundable “shipping and handling fee.”(Davis et al., 1998; Liu & Wei Kwok, 2003; Seo, Yoon, & Vangelova, 2016). Policies that do not impose monetary restrictions are considered as more lenient. 3) Effort leniency: Some e-retailers create some “ hassles” for product returns regarding

(14)

14

Return policies with a greater scope of “return worthy” are considered more lenient. 5) Exchange leniency: While some retailers accept cash refunds, others only allow store

credit or product exchange for the returned item (Davis et al., 1998). Return policies that allow case refunds be considered more lenient.

Three theoretical mechanisms are supporting that a lenient return policy has a positive effect on product purchase. First, signaling theory has been put forth to explain how return policy act as positive quality signals by the E-com retailer (Wood, 2001; Boulding 1993). Second, consumer risk theory can be applied to suggest that return policies should reduce the financial and product risk that consumer’s fee; before product purchase (Van de Poel and Leunis, 1999). Third, construal level theory is used as a basis for posting that individuals faced with lenient return policies are likely to focus on the benefits of purchase rather than the cost of purchase (Janakiraman and Ordonez, 2012). These three theories have a common thread to postulate that lenient return policies can lead to an increase in product purchase. Moreover, some research argues that customers complain that increasing restrictiveness of returns policy could bring more risks on the purchase, which can diminish customers’ satisfaction and loyalty on the online store (Passy, 2002; Haarlander, 2001).

On the other hand, Lee (2001) suggests restricting the return deadline, up to a 14-day return policy in the fast-fashion apparel industry, due to the most contentious category driving high volumes (up to 30%) returns and limited time to resell. Some researchers also state that a lenient return policy can stimulate customer to order the items in different sizes and colors or even order products that do not fit his or her needs, and ultimately leading to higher return rates (Ryder, 2010; Reinartz et al., 2002; Su, 2009; Grewal et al.,2004). Notably, in the clothes and shoe section, product return rate is even more highly influenced by the return policy (Van Coolwijk, 2014). An experiment conducted by Petersen and Kumar in (2010) in a fast-fashion industry shows the return rare under the leniency policy about the scope is dramatically higher than a strict return policy. However, Freling (2016) in his study he finds that leniency in time can reduce return rates which are different from what the other authors say. He explains this result as an endowment effect, which suggests that the longer consumers process a product, the more attached to it and become less likely to return them.

(15)

15

profits.

Free shipping promotion

Nowadays, online retailers have implemented various shipping promotions, which can be classified into three categories (Dinlersoz & Kok, 2006);

1) Unconditional free shipping (UFS): Online retailers are in charge of all shipping costs for all orders. Which means, customers can experience free delivery at home without any restrictions.

2) Contingent free shipping (CFS): By far the most common CFS is the free shipping if their order value equal or larger than a minimum order threshold (Barry, 2010). There is also another form of CFS, like Amazon’s Prime, only loyal, profitable customers can be rewarded with totally free shipping.

3) Non-free shipping policy: Based on the size and/or weight, certain merchandise incur a shipping surcharge. The company will not include this surcharge the customer should pay it by themselves based on the freight rate provided by different carrier company.

From the survey conducted by earlier researchers, it illustrates free shipping is a factor to help prevent online shipping abandonment at checkout time, especially for the less affluent users who are coming online in increasing numbers (Kukar-Kinney & Close, 2010). However, the benefits of increased revenue are countered by multiples costs. One, the retailer has to bear the freight costs to offer free shipping. Second, the retailer has to face the increased return rate and also higher costs on processing returns. For this case, Leng (2010) and Boone (2013) propose that the free shipping must be presented to the customers only after a certain purchase amount is reached so that margins on the products shipped can be high enough to cover the expense of shipping the item. Consequently, they think CFS policy is the most effective promotion in increasing the revenues of the online retailer. However, offering free return policy can encourage customers to order quickly, not thoughtfully through a real e-customer behavioral study from Lantz (2013). Evidence states that when customers are aware of a free product return, they are more likely to purchase, along with the growing frequency of returns (Seo et al., 2016). Whereas, when a customer realize that they need to pay for every return or the level of effort to go through a return is high, he or she is probably doing his or her best to order the desired product in one go and try to avoid returns (Jack & Powers, 2004).

(16)

16

3. Methodology

The project aims at identifying factors in the return product flow that cause costs, organize these factors in a quantitative framework, and to test this framework in an empirical setting. In line with the objectives the process is organized into three phases 1) Case study to identify cost factors; 2) Create a cost framework; 3) empirical testing of the framework. Below, it shows the different research method in different phases. Lastly, the data validity is discussed.

3.1 Phase 1: Case study to identify cost factors

The primary goal in this phase is to identify cost factors based on the case study. From the previous literature study, six key processes (Returns acquisition, transport, inspection, disposition, remanufacturing and marketing) are known to generalize in the reverse supply chain (Bowersox, 1991; Chan 2008; Li, 2008). However, how to allocate costs to each process is unknown. The author also doubts that in a real world more processes are included beginning from customer deciding to return an item and ending until customers get the refund. Under this consideration, the author does a case study to investigate an integrated chain and specific activities inclusive of processing product returns in practice. In a fast-fashion industry, on the one hand, time leniency return policy can impair reselling opportunity (Lee, 2001); on the other hand a lenient product return policy and free shipping promotion can increase customer’s likelihood to purchase a product in the first place (Van de Poel et al., 1999; Janakiraman, 2012; Passy, 2002; Baylws, 1998). Also, some literature illustrates that lenient return policy and free shipping promotion are two factors to cause higher returns (Wood, 2001; Seo at al., 2016), while Freling (2016) state that more extended return deadlines may help the retailer to curb returns because of the endowment effect. Extant scholarly research offers no clear-cut answers on the trade-off between sales and product return impacted by lenient return policy and free shipping promotion. Aiming to fix this gap, a case study can be done to investigate what factors cause these trade-offs. Moreover, other factors to influence the cost framework on product returns can also be discovered during the case study.

(17)

17

In order to meet all the aforementioned criteria, Crocs. Inc, a footwear manufacturing company, is selected as the overall research context. This is because of footwear E-commence with a leading high product return ratio due to lacking fit and trial elements (Mollenkopf et al., 2011; Passy, 2002). The average rate of return for footwear reaches 30%, which is much higher than other general merchandise sit in the average 5-15% range (Lee, 2001). Apparently, online footwear retailers today face a challenge with product returns, consequentially more problem about costs of product return emerge in the company. Furthermore, managing and reducing the increasingly expensive on “reverse supply chain” has become significant for Crocs. For this reason, Crocs wants to find out all split in costs generated from the “reverse supply chain.” At the same time, a customer from Crocs expects to be offered free shipping and lenient return policy. However, these two offers have a trade-off between sales and returns. Crocs does not know whether this will affect its bottom line on financial results. Therefore, Crocs would like to investigate how return policy and free shipping affect their net profit. In this way, Crocs have the same objective with this research; it is a most suitable case company for this research. Crocs was initially launched in the US, and then expanded to Europe and Asia. It is estimated that over the last 13 years, Crocs has already sold more than 300 million pairs of shoes. This research study is conducted in Crocs European Head office, Hoofddorp, The Netherlands, which manages the European business. Currently, all online purchased orders in Europe are delivered from the warehouse in Rotterdam, and when product return occur they also need to be shipped back to Rotterdam for quality checking, restocking and repackaging.

Data Source:

In this phase, data can be collected from two steps. Step I: to investigate the overall product return process, and identify the cost components of product returns. Step II: to investigate what factors influence sales and returns caused by return policy and free shipping. The data source for each step is present below.

Step I: Cost components of product returns

(18)

18

1 below.

Department Personnel

Supply Chain Planning Sr. Director of Operations

Customer Service E-com customer service manager

Warehouse Warehouse Manager

Warehouse Transportation Manager

Table 1: List of the department and personnel interviewed in Step I

Step II: How return policy and free shipping affect sales and return costs

In this step, data is firstly collected from the author’s physical experience. Before this research, the author did a six-month internship in E-com department in Crocs. From the internship, she gains a better understanding of E-com context based on personal observation and some informal conversations with her coworkers. This is beneficial to enrich additional field notes during the interview (Eisenhardt, 1989).

From the literature, it is known that return policy and free shipping are two main indirect factors to influence the cost of product returns. To develop a full-scale image on how these two factors influence the cost of product returns, the researcher was seeking for someone equipped with professional knowledge on E-com. He should have a comprehensive overview of KPIs and detailed information about product returns, return policy and free shipping. Furthermore, E-com sometimes conducts some A/B tests (Sometimes called split testing, is comparing two versions of a web page by showing the two variants to a similar visitor at the same time) to measure the performance under different scenarios. Therefore the author is looking for a candidate who can access to the various test results and have the ability to analyze all E-com data. Through a candidate scanning, an E-com data analyst frequently does tests on the website and is a master of analyzing and evaluating all kind of data every day in E-com. Hence he is a perfect person who met all requirements in this step.

(19)

19

Data Analysis:

Once all the interview and transcription work is done, a series of data analysis processes are carried out. Coding is one mainstream method to analyze textual data and is selected to help the researcher to determine patterns, categories, linkages and relationship between variables (Basit, 2017). In this research, cost components of product returns from the literature are used as the initial pre-defined code list and then expanded during the interviews to capture emerging themes. Both inductive and deductive coding is used with the help of the software package ATLAS, ti,.which can contribute to more systematic analysis procedures and guard against information-processing biases. The specific coding process is carried out as follows:

- First order coding: The first category infers the quotes, sentence, and paragraphs from the previous literature and the actual words and phrases used by the respondents, and then uses labels that are optimal to describe a phenomenon mentioned in the text. - Descriptive coding: First order codes are grouped and translated into descriptive codes. - Interpretive code: Pooling descriptive codes to integrate and classify a central

phenomenon identified to response each aspect of costs of product returns. And then the interpretive code is assigned to one of the three main topics which are direct factors of product returns, time leniency in return policy and free shipping.

The table below provides the methodology of the coding tree.

Table 2: Overview of Coding Tree

Descriptive code (Sceond order code) Interpretive Coder (Third order code) Main Factors

Freight cost Costs of carrier to collect returns and transport to WH Labor cost on quality check

Repackage fee Handling fee Restorage fee Labor costs Overhead costs

Units return pattern on days Defected rate

MAC cost for defected items EOL units

Reselling price Changes on Units Sold Retail price Purchase orders Shipping cart abandonment Threshold

Conversion rate for different threshold AOV for different threshold

Shipping fee prepaid in forward logistical Returned orders

Returned orders which meet threshold Shipping cost lose for CFS

Direct factors to cause costs of product returns Customer service cost happen on processing product

returns Costs in DC process

Cost lose casued by defected units Product depreciation due to the seasonality shifted

Sales benefits form lenient return policy

Time leniency in return policy

Sales benefits form CFS Sales benefits form NFS

Shipping cost lose for NFS

(20)

20

3.2 Phase 2: Create a cost framework

According to the coding tree from phase 1, a cost framework can be created as two steps in phase 2. Step 1, building a framework includes all cost components to process a product return. Then in step 2, it is aimed to build a framework to evaluate the change of financial performance by adapting the shipping and return policy. Therefore, the framework built in step 2 considers the factors of cost of product returns as well as sales influenced by return policy and free shipping promotion.

In this phase, ABC is the main method to be applied in step 1. Four steps are shown to implement ABC;

1) Identify activities, from the coding tree, to identify what specific activities to process a product return

2) Assign resource costs to activities: to understand what cost, is the labor cost, freight cost or overhead costs are involved in each activity, and what the unit is applied to each cost 3) Identify outputs: the outputs in this framework are identified as the total returned units 4) Assign activity costs to outputs: A framework about costs for total returned units can be

built after assigning activity costs to the total returned units.

3.3 Phase 3: empirical testing of framework

After the cost framework is built, phase 3 is applying the real data collected from the company Crocs to empirical testing the framework. In this phase, flexible process and rich data collection are required to provide a better understanding and statistical analysis of the underlying research question (Voss, 2009).

To complete the testing, further additional data, reports and other archival information are regarded as the reference in this phase. The documents of the company accessible to the researcher and relevant to research are reviewed and are given in the table below.

Document type Description

2017 Q4 financial cost report This report includes all activity based costs for each

department from 1st

Oct to 31st

Dec 2017.

2017 Q4 E-com return dashboard From this document, the return pattern from 1st

Oct to 31st Dec 2017 can be studied. Meanwhile, the defected units on days level are also included in this report.

2017 E-com customer service staff model This data is applied by customer service and used to calculate the weekly time spent on e-mail, chat, and phone for each website of Crocs in 2017. So studying this document, the author can know the exact time that the staff in CS spent on responding to customers.

2017 Customer service e-mail subjects This document is downloaded from the Crocs Moxie

(21)

21

subjects, then how many e-mails are related to the subject returns can be found.

2017 RMA data This data is used by DC to record all information, time, and

labor cost for each activity to process returns spent in the RMA department.

2017 A/B test on free shipping for E-com This data is used to test how free shipping affect E-com performance, include sales and returns. Data shows traffic, customer purchase details (purchase order number, purchase date, AOV, unit items, purchase price), returned items information (returned orders, returned units, returned date), and the threshold for each website of Crocs. This data is available for the testing period, from 15th Sep to 15th

Oct. Table 3: The documents read as part of research at Crocs for Phase 3

3.4 Data Validity

(22)

22

4. Case Study

Based on the data coding from the interviews, several findings could be found that hold significant value for preparation of building cost framework. The following sections focus on the findings of the cost components of the product returns, and findings about the return policy and free shipping. The final section discusses what findings from interviews align with the literature, and what parts are missing in theory.

4.1 Findings

During the interview, to answer the underlying research questions, the findings can be separated into two aspects. One, is the finding about the cost components, which can be summarized from the interview by comprehending a full-scale product return flows. Second, is the finding about the factors that can cause the trade-off between sales and costs due to the return policy and free shipping.

Findings about Cost components

In the real world, processing product returns are more complicated than the theory. Hence more costs are involved in calculating the actual money spent on return items. Director of Operations in Crocs mentioned that “the product return starts from the customer, and end until customers received their refunds and advised to divide the product returns process into three processes.” The three processes are reverse logistic, DC process and Customer Service and are explained below-

1) Reverse logistic - Some online retailers provide free returns, but others require customers themselves to ship and pay the returns back to the warehouse, “Normally, for those free returns, to reduce costs, company may corporate with the third reverse logistical group, that are responsible to collect all return parcels together and ship directly to the central locations which are warehouse or central distribution centers. For those free returns, the company need to pay the shipping costs, that is the only component included in this process.” the transportation manager, working in a warehouse.

2) DC process - A warehouse manager said: “DC processes in a footwear industry on product returns can be executed as Inspection, Disposition, and Restorage.” From the interview, applying ABC to allocate each cost and the cost components can be determined as;

- Labor costs in inspection activity: It includes the time staff takes to verify information of received returns with warehouse management (WM) system and quality check on them

(23)

23

- Handling costs in disposition activity: It comprises of all the labor costs involved in movement of stock from one place to another place

- Restorage fee: These are charged for storing the return items

3) Customer service: The service in product return process helps customers to deal with all questions related to returns. “The questions could be asked like return inquiry, where to return, return label created, where are my returns, when can I get my refund. etc.,” Customers can use e-mail, call and chat to get touch with customer service team. “To calculate costs in this part, the cost can be split into two parts. First is the labor costs allocated to e-mail, call and chat reply that is related to product returns. Another is the overhead costs to assume how much percentage of activities in customer service is related with the product returns, then applying the assumption weight to calculate the overhead cost of product returns” said the customer service manager.

Overall a completed cost components list of product returns can be summarized in table 4 below

Table 4: Cost components for cost framework

Findings about effects of return policy and free shipping

From the interview, “Depending on whether it is a durable good or a consumable good, whether it is high-fashion or fast-fashion, those different segments of the market have different reasons for buying and they have different concerns for risks and quality. Therefore, different leniency in return policy plays a different role in sales and product returns. For fast-fashion, like a footwear industry time leniency is a significant factor to influence the cost of product returns” by Warehouse Manager. For the whole interview, only time leniency from the return policy is considered a factor that affects the cost of product returns.

Therefore, findings of how time leniency and free shipping promotion affect sales revenue and costs of product returns can be summarized in table 5.

Factor Effects on sales revenue Effect Interviewers

Time Leniency Due to the time leniency in the return policy, some items are already out of season when they are returned, especially in a fast-fashion industry. So the company has to clear them with a low price. Sometimes even less than their cost.

Negative Eugen; Martin

Aspects Activities Cost components

Reverse logistic Collect returns and transport to WH Freight cost DC process

Inspection Labor cost of quality check

Disposition Repackage fee

Handling fee

Storage Restorage Fee

(24)

24

Free shipping 2

Conversion rate varies with the change of threshold. Based on company’s A/B test3

, they found that conversion rate decreases with the increase in the value of the threshold.

Positive Eugen

Average purchase value (AOV) grows with the increase of the value of threshold

Negative Eugen

Factor Effects on costs of product returns Effect

Time Leniency The longer time allowed to return items, the more defected items are received by the warehouse. This leads to a loss in costs of products

Negative Martin

Lenient time in return policy may increase the product return rate and the possibilities of product return fraud.

“After 60 days, there is a high rate of the returned shoes in a worn condition.”

Negative Martin

Free shipping Return rates vary with the change of threshold. When increasing the value of threshold, the return rate grows as well

Positive Eugen

Table 5: Interview Summary about effects of time leniency and free shipping 2

Conversion rate: The number of customers who have completed a transaction divided by the total number of website visitors.

3

A/B test: A controlled experiment with two variants, A and B. Crocs uses it to test how web performance change under the different threshold.

4.2 Discussion

(25)

25

5. Cost Framework

Based on the interviews done at the case company and the parameters mentioned that affect the costs, the framework is built in this chapter. The chapter starts with formulating cost components for product returns. Further, two factors return policy and free shipping are added in these cost components to observe the consequences on financial results. The chapter ends with the summary about the cost framework of product returns.

5.1 Model development

To develop the cost model about product returns, two steps are discussed below. In the first step, it reveals all activities which are concerned with the reverse supply chain; then in accordance with these activities, each cost can be allocated. After the cost framework is built, how other two factors, return policy and free shipping promotion, affect the cost framework built in step I is discussed in step 2.

Configuration of Cost Framework based on the product returns process:

According to the findings of cost components in chapter 4, the cost framework can be built in table 6 below, and an explicit explanation for each variable is given below the table. Before building the model, some assumptions are made which are as follows-

Assumptions:

- No remanufacture, repair or recycle involved in any returned items for any purpose. - Each returned items are required to repackage and are not kept in their returned package - The working time to allocate to each labor costs in this framework are assumed to be equally

divided to each worker in the same activity

Table 6: Setting of cost components 1

N: is the total units of product returns

Aspects _componentsCost Unit cost Units _{for returns}Total cost

Reverse

logistic C1 Freight cost Cost per unit T1 T1*N1

DC

process C2

Labor cost on

quality check Labor cost per hour L2i

Time to inspect a unit

(hour) t2i L2i* t2i*N

Repackage

fee Cost of package material per unit P2 P2*N

Handling fee Labor cost per hour L2h Average time to handle a _{unit (hours)} t2h L2h* t2h*N

Restorage Fee Storage fee per unit S2 S2*N

Customer service C3

Labor costs Labor cost per hour L3 Average time per contact _{for returns (hours)} t3 L3* t3 Overhead

cost

h3 Percentage of daily

activity is related with product returns

(26)

26

Based on the parameters set in table 6, the model about cost of product returns can be quantified as; the explicit explanation for each parameter are shown after the formula

Cost of product returns (Ca) = C1+C2+C3

= T1*N + N (L2i* t2i+ P2+ L2h* t2h+ S2)+ (L3* t3+ h3* r3) Reverse logistic (C1):

T1: Transportation costs per unit paid by the company to collect returns and directly ship to the

central location. If a company has no free return policy, which means customers need to ship their returns by themselves, then no transportation costs are taken in this part. T1 may vary

according to the size of return units; the more units company collect from the customer, the cheaper transportation cost per unit will be paid.

DC fulfillment cost (C2):

L2i: Labor cost per hour for those staffs working in the quality assurance area in the WH. They are

mainly responsible for checking if the returned items match with the WH return system. This is to check if the items are returned in good condition, like no damage, clean and product tags still on.

t2i: Average time spent on each return unit’s quality check.

P2: Repackage cost per return unit

L2h: Labor cost per hour for those staffs who are responsible for moving checked return units from

return area to back to the stock

t2h: Average time per unit spend on handling those checked return units back to the stock

S2: Storage fee per return unit.

Customer Service (C3):

L3: Labor cost per hour for customer service staffs.

t3: For each conversation with the customer, the average time to answer product return questions

that customer service staffs spend on e-mail, phone, and chat.

h3: Total overhead cost, including administration fee, office expense. etc

r3: The percentage that the total time spend on dealing with product return cases accounting for

(27)

27

Configurations of cost model under different scenarios:

According to the findings from literature and interviews, a new model can be developed when considering different configurations of the return policy and free shipping promotion. Before building the framework, some assumptions and factors are explained below.

Assumptions:

- All returned items can be resold soon, and there is no risk of staying in the warehouse forever

Factors:

- All benchmarks vary with the change of return deadline and free shipping promotion.

- Under EU rules, customers have the right to cancel and return their orders within 14 days for any reasons and without justification. It assumes that the benchmark of purchase orders when under 14 days return policy and non-free shipping is S*

- Under the same conditions in the last factor, the benchmark of total returned units is N*

- Benchmark of defected units from the all returned units is: D*

- Benchmark of depreciated units among all the returned units is W*

- The average selling price (ASP) for normal products is: Pa

- The cost for each returned unit is defined in phase that is: Ca

Based on all assumptions and factors made, the framework about the value of sales change due to the return policy and free shipping promotion can be quantified below, all parameters in this formula are shown in table 7

Revenue (R)

=Regular Revenue from sales – Cost lost in defected items – Depreciations on sales = S*

(1+r1)(1+r2)a - D*(1+rd)m- W*(1+rd1)(pa-pm)

Table 7: Framework about the revenue

Factors Sale value

Defect change rate rd

Mac cost m

change rate for

depreciated items rd1 Average markdown

price pm

AOV a

The change rate for the purchase orders rs2 Return policy

S*_a(1+r1)(1+r 2)

Cost D*_(1+r d)m

Sale lost from

Depreciations W*(1+rd1)(pa-pm)

Sales

Sales units Sale lost

Free shipping

(28)

28

All variables in table 7 can be described as-

rs1: the increased/decreased rate of the purchase orders, it caused by the change of time deadline

from the return policy

rd : the increased/ decrease rate of the defected returned units, it caused by the change of time

deadline from the return policy

m: the average MAC cost for returned units, MAC cost is the cost of the products that occurred from factory to the warehouse

rd1: the increased/ decrease rate of the depreciated items among the total returned units, it caused

by the change of time deadline from the return policy

pm: the average markdown price for those depreciated items

a: average customer order value, it varies with the change of threshold in the free shipping promotion

rs2: the increased/decreased rate of the purchase orders, it caused by the change of free shipping

promotion

When changing the return policy and free shipping promotion, the model about cost of product returns can be formulated as below, and all parameters are shown in Table 8

Cost of product returns (Cb)

= Cost of product returns +Shipping costs of returned items prepaid in forward logistic = CaN*(1+r1)(1+r2)+t1ut

Table 8: Framework about returns

All variables in table 8 can be described as

r1: the increased/decreased return rate and is caused by the change of time deadline from return

policy

r2: the increased/decreased return rate and is caused by the change of free shipping promotion

ut: the number of orders which reach the threshold are returned. So shipping costs on these

orders are lost, assume customer return all units per order.

t1: shipping cost per parcel, this happens in forward logistic. Because the shipped items to the

customer are returned, the shipping cost paid by the company is considered as a cost lost due to the product returns. Assume each purchase order only has one parcel.

Factors Returned units FL shipping _Cost

Return policy

The change rate for the returned units r1

The orders which reach the threshold are

returned

ut Shipping cost per parcel t1 N*_(1+r

1)(1+r2)

t1ut

Returned order for free shipping Returned units

Returns

(29)

29

5.2 Summary

From above sections, it is known that time leniency can affect customer purchase behavior to increase sales. Meanwhile, it can lead to more costs on defected items and revenue loss on those out of season products from lagging returns. For free shipping promotion, on the one hand, it can stimulate customers to check out their online shopping cart to increase conversion rate, and on the other hand, the return rate will be high due to this policy. Additionally, for CFS another form in the free shipping promotion, the AOV, conversion rate along with the return rate are influenced by the threshold

Consequently, the new model can be formulated as with some assumptions;

Assumptions:

- No remanufacture, repair or recycle involved in any returned items for any purpose. - Each returned items are required to repackage and are not kept in their returned package - The working time to allocate to each labor costs in this framework are assumed to be equally

divided to each worker in the same activity

- All returned items can be resold soon, and there is no risk of staying in the warehouse forever

The Expected Profit E(p)

= Revenue (R)-Cost of Product Returns (Cb)

= (Regular Revenue from sales – Cost lost in defected items – Depreciations on sales)-(Cost of product return +Shipping costs of returned items prepaid in forward logistic)

= [S*

(1+r1)(1+r2)a - D*(1+rd)m- W*(1+rd1)(pa-pm)]- [CaN*(1+r1)(1+r2)+t1ut]

(30)

30

6. A costing model test with Crocs

In this chapter, Crocs is applied to test the model. In order to verify the validity of this costing model, product returns process is identified firstly, and then mass data is collected to test the model. Finally, through a model testing, a conclusion about the company is given.

6.1 Product returns process in the Crocs

This section entails the comparison and similarities between the model developed as an outcome of the research to the case company Crocs.

Similarities between framework built and process done in Crocs

By conducting several interviews in the company, a product returns process which is matched with the model built in Chapter 5, can be identified into five processes (see figure 1 below). Detailed activities in each process are explained afterward.

Figure 1: Crocs product return process

1) Customer request a return: When customers are not satisfied with the received items, they decide to return items.

2) Sometimes, when customers have any questions about the product returns, they can contact customer service team of the Crocs by phone, chat, or e-mail. Some sample questions are “How to return my shoes”; “Where can I find my return label”; “Where is my return items”; “When can I get my refund” etc.

3) Reverse Logistics

4) DC process, the whole DC process regarding product returns is shown in the flowchart below. When returned items are shipped back to the warehouse, they all then go to the RMA department to start processing returns. The costs, therefore, start from here.

CONFIDENTIAL

(31)

31

Figure 2: Crocs DC process about product returns

Figure 3: Warehouse layout in Crocs

(32)

32

Differences between framework built and process done in Crocs

This section mentions those data are failed to collect from the Crocs 1) Overhead cost in customer service:

2)

3) No storage cost is provided by Crocs, so this data is failed to collect

6.2 Testing the model

Step I: Test in the basic cost framework

In this phase, since data before September for Customer Service are missing due to the system update. All data applied in this model is from September to November, three months in total. To apply data to test the model in Step 1, some assumptions and factors are shown firstly

Assumptions:

Factors:

- Using time per conversation for product returns to calculate labor cost on processing returns, this is also the cost for customer service cost (see appendix B)

Table 9: Summary about cost components on product returns

CONFIDENTIAL

(33)

33

Table 10: Summary of product returns costs

Step II: Test in the model under different scenarios

In this phase, the time frame is from 15th_{Sep to 15}th_Oct,

Assumptions

:

-

When there is no change in the return policy, the defect rate % among returned units for each threshold is same with the rate shown in Appendix C

-

When there is no change in the return policy, the rate for warehouse sales among the returned units are same with the rate we calculated in Appendix C

- Every purchase order has only one parcel shipped

- Customers returned all the bought items. Situation like order more items but just return one pair are not considered in here

Factors:

In accordance with these assumptions, the profit under different scenarios can be calculated in the table below;

CONFIDENTIAL

(34)

34

Table 11: Profits after calculating costs of product returns under different scenarios

(35)

35

7. Conclusions

The primary intent of this paper is to quantify the cost components of product returns in an E-com footwear industry by generalizing the product returns flow. And simultaneously to investigate how return policy and free shipping promotion can be designed to maximize profits. To achieve these goals, the author was first to use ABC method to allocate costs to each return activities emerging in reverse logistics, DC process and customer service. Then, based on the cost framework of product returns, to build a financial consequence model when adding two factors, free shipping, and return policy. Specifically, the model considers the case of the time leniency in return policy, that is the different return deadline applied in the return policy can lead to different purchase behavior and return behavior. Through the interview together with the literature review, it concluded that the sales units, defective units, reselling opportunities and return units all might be affected by the return deadline. Additionally, the model also considers the case of the UFS and CFS in free shipping promotion. To consider for the increased revenue potentially generated by the promotion, the retailer also needs to consider the returns and value lost on shipping to customers. Therefore, the final model encompasses cost of defective units, product returns, shipping to customers, sales lost on product depreciation and sales revenue of the purchase orders. Once, these numbers are figured by the company; the model can present the optimal return deadline and a threshold value that maximize profits.

This model has many implications. First, it fixed the gap on missing research about costs of product returns. Future study can extend cost of product returns' research based on this model. Second, it gives a method for a retailer to implement a data-driven model to computer optimal return deadline and threshold level. Third, the retailer can design some parameters in this model to achieve a certain level of profit – this could be finding a way to minimize the costs associated with shipping, processing or labor costs. For example, cooperating with the third party to get lower price with freight consolidator to save costs on transportation (Guide et al., 2000)

7.1 Limitations and future study

While this research makes contributions to the theoretical and practical filed, it is fraught with some limitations. First, as this research has an explorative character, naturally some limitations exist in this early stage of theory generation. Especially conclusion from the chosen context, single case study, and company-specific created artifact might not yet be completely generalizable. Further, although most data input can be collected from the case company to empirical test the model, there still has some missing data. Therefore, it cannot be completed verify the validity of the costing model. So, in the future, do a multiple case study in different companies and collect more integrated data can generalize the situation better and improve the validity of the data.

(36)

36

deadline of return policy, will the result sill be aligned with the assumptions? For this doubt, to improve the reality and accuracy of the research study, in the future the author can experiment to test the website and return performance by setting different restrictions on return deadline. What’s more, this study only investigates the time leniency in return policy, further study might examine other four leniency factors monetary leniency, effort leniency, scope leniency and exchange leniency (Davis et al., 1998). For example, Petersen and Kumar (2010) suggest that scope leniency in return policy has a positive effect on return behavior.

Third, some assumptions are made to build the model, which may limit the applicability of the model. In the future, it would be useful to develop a new method to address this research problem without these assumptions.

(37)

37

References

Allott, A. (2004, Jan). Activity Based Management can work for your company. CIMA Insight. Barrett, R. (2005, Feb). Getting a better view of business with activity based costing. CIMA Insight. Basit, T. (2017). Manual or electronic? The role of coding in qualitative data analysis. Educational

Research, 45(2), 143-154.

Blanchard, D. (2005). Moving forward in reverse. Logistics Today. Logistics Today , 1-8.

Bonifield, Carolyn, Cole, C., L, R., & Schultz. (2002). Product Returns on the Internet: A Case of Mixed Signals? working paper. University of Vermont, Marketing Department.

Boone, T., & Ganeshan, R. (2013). Exploratory analysis of free shipping policies of online retailers.

International journal of production economics, 143(2), pp. 627-632.

Boulding, W., & Kirmani, A. (1993). A consumer-side experimental examination of signaling theory: do consumers perceive warranties as signals of quality? Journal of Consumer Research, 20(1), pp. 111-123.

Bower, A., & Maxham, J. (2006). Customer Responses to Product Return Experiences, working paper.

University of Virginia, McIntire School of Commerce.

Bowersox, J. D., & Daugherty, J. P. (1991). Achieving and maintaining logistics leadership: logistics organisations of the future. Logistics Information Management, 4(3), pp. 42-47.

Bunn, J. (1999, Aug). Centralizing reverse logistics: How to understand if it will work for you.

Cai, X. (2012). The Study of Return Reverse Logistics in B2C E-commerce Transactions.Logistics.

Engineering and Management, 32(4), 106 - 108.

Chan, S. F., & Chan, K. H. (2008). A survey on reverse logistics system of mobile phone industry in Hong Kong. Management Decision, 46(5), pp. 702-708.

Charlton, G. (2007, March 15). Returning online purchases is too much hassle – survey.

Che, Y.-K. (1996). Customer return policies for experience goods. Journal of Industrial Economics, 44(1), 17-24.

Cutcheon, D. M., & Meredith, J. R. (1993). Conducting case study research in operations management.

Journal of Operations Management,, 3, pp. 239-256.

Daugherty, J. P., Ellinger, E. A., & Gustin, M. C. (1996). Integrated logistics: achieving logistics performance improvements. Supply Chain Management, 1(3), pp. 25-33.

Davis, S., Hagerty, M., & Gerstner, E. (1998). Return policies and the optimal level of Bhassle. Journal of Economics and Business, 50(5), 445-460.

Eisenhardt, K. M. (1989). Building theories from case study research. The Academy of Management Review, 4, pp. 532-550.

Fleischmann, M. (2001). Reverse logistics network structure and design. Rotterdam, The Netherlands: ERIM Report Series Research in Management.

Freling, R. (2016, Jan 19). Researchers examine effect of return policies on consumer behavior. Friedman, A. L., & Lyne, S. R. (1995). Activity-based techniques: the real life consequences. London:

CIMA Publishin.

Ganeshan, R., Ring, L., & Strong, J. (2004). Centralizing Retailing: How much Supply Chain Costs is it Worth? Proceedings of the International Seminar on Production Economics, 139-153. Ginger, K. (1999). Internet-only retailers struggle to improve product return process:One solution to

returns headaches is to bring in third-party ‘reverse logistics’ provider.

Gooley, T. B. (1999). There and back again. Logistics Management and Distribution Report.

Grewal, D., Iyer, G., & Levy, M. (2004). Internet retailing: Enablers, limiters, and market consquences.

(38)

38

GriffisES., RaoS, GoldsbyJT., & NiranjanT.T. (2012). The customer consequences of returns in online retailing: An empirical analysis. Journal of Operations Management, 30(4), 282-294.

Haarlander, L. (2001, September). Diminishing returns: target, home depot tighten return policies. The Buffalo News.

Heiman, A., Williams, B., & Zilberman, D. (2001). Demonstrations and money-back guarantees: market mechanisms to reduce uncertainty. Journal of Business Research, 71-84, 71-84.

Jack, E. P., & Powers, T. L. (2004). Volume flexible strategies in health services: A research framework.

Production and Operations Management, 13(3), 230 - 244.

Janakiraman, Narayan, & Ordó˜nez, L. (2012). Effect of Effort and Deadlines on Consumer Product Returns. Journal of Consumer Psychology, 22(2), 206-271.

Karlsson, C. (2009). Researching Operations Management. New York : Routledge.

Kukar-Kinney, M., & Close, A. G. (2010). The determinants of consumers’ online shopping cart abandonment. Journal of the Academy of Marketing Science, 38(2), pp. 240-250.

Lantz, B., & Hjort, K. (2013). Real e-customer behavioral responses to free delivery and free returns.

Electronic Commerce Research, 13(2), 183-198.

Lee, C. (2001). An analytical framework for evaluating e-commerce business models and strategies .

Internet Research , pp. 349-359.

Lewis, M., Singh, V., & Fay, S. (2006). An empirical study of the impact of nonlinear shipping and Handling Fees on Purchase Incidence and Expenditure Decisions. Marketing Science, 25(1), 51-64.

Li, X., & Olorunniwo, F. (2008). An exploration of reverse logistics practices in three companies. Supply Chain Management: An International Journal, 13(5), pp. 381-386.

Liu, X., & Wei Kwok, K. (2003). An empirical study of product differences in consumers’E-commerce adoption behavior. Electronic Commerce Research and Application, 229-230.

Lutters, W., & Seaman, C. (2007). Revealing actual documentation usage in software maintenance through war stories. Information and Software Technology, 49(6), pp. 576-587.

Marien, E. J. (1998). Reverse logistics as competitive strategy. Cahners Business Information.

Meelah, R., & Ibraham, D. (2007). Factors influencing activity based costing (ABC) adoption in manufacturing industry. Investment Management & Financial Innovations, 4(2), pp. 113-124. Mollenkopf, D., Frankel, R., & Russo, I. (2011). Creating value through returns management: Exploring

the Marketing-Operations interface. Journal of Operations Management, 29(5), 391-403. Mukhopadhyay, S. K., & Setoputro, R. (2005). ptimal return policy and modular design for

build-to-order products. Journal of Operations Management, 23(5), 496-506.

O’Neill, S., & Chu, J. (2001). Online Returns Management Strategies. IBM Global Services eStrategy, 18.

Passy, C. (2002, Nov). Cranky consumer: get set for many unhappy returns.

Petersen, A., & Kumar, V. (2009). Are product returns a necessary evil? antecedents and consequences.

Journal of Marketing, 73(3), 35-51.

Petersen, J. A., & Kumar, V. (2010). Can product returns make you money? MIT Sloan Management Review, 51(3), 85.

Poel, V. d., & Leunis, D. a. (1999). Consumer Acceptance of the Internet as a Channel of Distribution.

Journal of Business Research, 45(3), 249-256.

Reinartz, W., & Kumar, V. (2002). The mismanagement of customer loyalty. Havard Business Review , 86-94.

(39)

39

Rogers, D. S., & Tibben-Lembke, R. S. (2001). An overview of reverse logistics practices. Journal of Business Logistics, 22(1), pp. 22-28.

Ryder. (2010, June 9). Reverse logistics–From black hole to untapped revenue stream.

Seo, J. Y., Yoon, S., & Vangelova, M. (2016). Shopping plans, buying motivations, and return policies:impacts on product returns and purchase likelihoods. Marketing Letters, 27(4), 645-659.

Srivastava, K. S. (2008). Network design for reverse logistics. Omega, 36(4), pp. 535-548.

Stock, S. J., & Shear, H. (2002). Many happy (product) returns. Harvard Business Review, 80(7), pp. 16-17.

Su, X. (2009). Consumer returns policies and supply chain performance. Manufacturing & Service Operations Management , pp. 595-612.

Thomason, E. (2013). Online retailers go hi-tech to sizeup shoppers and cut returns. Reuters. Retrieved. Turney, P. B. (1989). Using Activity-Based Costing to Achieve Manufacturing Excellence. Journal of

Cost Management, pp. 23-31.

Voss, C. (2009). Case Research in Operations Management. Researching Operations Management, pp. 162-195.

Wood, S. L. (2001). Remote Purchase Environments: The Influence of Return Policy Leniency on Two-Stage Decision Processes. Journal of Marketing Research, 38(2), 157-169.

Yin, R. K. (1994). Case study research: Design and methof. Newbury Park: CA:Sage Publications . Yin, R. K. (2009). Essential guide to qualitative methods in organizational research. In Case Study

Research: Design and Methods.

Empirical study on cost components of E-com product returns in footwear industry