Customer Segmentation Based on Buying and Returning Behaviour

In this article, we will be reviewing a research paper, titled: ‘Customer segmentation based on buying and returning behaviour’. The focus of this research paper is based on two aspects, which are: 

  • To determine if a general-purpose strategy conforms to the clothing businesses that involve e-commerce
  • To properly assess for certain if customer rates of return are a core element of the establishment of revenue growth and, if it is, then to explore the importance of returns management (RM) in the supply chain management.

The research paper was written by Klas Hjort, Björn Lantz, Dag Ericsson, and John Gattorna, and published in the International Journal of Physical Distribution & Logistics Management (IJPDLM) and was accepted on 18th March 2013. Hjort, Lantz, and Ericsson belong to the University of Borås in Borås, Sweden, whereas Gattorna is affiliated with the University of Technology, Sydney (UTS), Sydney, Australia, and S P Jain School of Global Management, Singapore.

The problem with customer returning behaviour

According to supply chain management (SCM) academics, supply chain differentiation (SCD) projects and ideas present supply chains that focus on client requirements. This leads to supply chain differentiation (SCD), which supposedly provides a way to boost supply chain management (SCM) efficiency. Although supply chain differentiation (SCD) is a hot topic among users, research has mostly ignored it, and that is because it cannot form an adaptive framework.

Supply chain management (SCM) has evolved from a functional optimization tool to a key tool for differentiating yourself from the competition. Customer needs are constantly subjected to variation and shifting, necessitating a unique supply chain management (SCM) strategy. Given all that, it is now a known matter that the universal technique to supply chain design (SCD) is no longer relevant, according to research (Beck et al., 2012, 213-239).

Methods used for this research

Results generated by sales and returns that were purely transactional were collected and utilized to classify consumers on the basis of their purchasing and returning habits, with total benefit rates calculated for each client.

In the fast-moving consumer goods (FMCG) industry, particularly in clothing e-commerce, creating a supply chain and strategic planning necessitates a thorough awareness of the responses clients will give based on their needs and requirements and the fulfillment of the associated expectations. That is why developing supply chain strategies have to:

  • Proximal to the competition
  • Have a particular context

These conditions implied the use of the information from a business. The study business,, was chosen largely for the aim of checking certain ideas, for the following reasons:

  • It did not categorize its consumers or distinguish its good or service offers to consumers. 
  • The company agreed to provide records to the study in order to evaluate the mentioned assumption at both the corporate and consumer stages used records from its Enterprise system for the quantitative survey. The data included all of their transactions from their market sectors in Sweden, Norway, Denmark, and Finland over the period 2008-2009 (2 years).   

The researchers used extensive computations to disclose many characteristics of every client’s conduct. These included:

  • The sales numbers for the consumer’s purchase
  • The returns data
  • The contribution margin 

The overall sales, total contribution margin, overall amount of orders, mean contribution margin, average sales per transaction, and the overall number of returns for every customer were then examined.

Customers’ monetary support was classified as per their purchasing and returning patterns to evaluate the hypothesis concerning the construct of effectiveness.  

  • Customers were divided into two groups: repeat consumers and non-repeat consumers.
  • Whether they are returners or non-returners (according to whether they completed just one or multiple acquisitions throughout the analysis period) 

Four distinct categories of clients developed from this viewpoint, which were labeled Types A-D as shown in Figure 1 (Table 1 in the original research paper) (Figure 1).

Figure 1: Basic Types of Customers

From a country-by-country perspective, variations in input per order, input per client, and contribution each year across the four types of consumers were characterized and then examined employing two-way ANOVAs. 

An ANOVA test is used to determine whether or not the findings from the survey or research are meaningful. To put it another way, they help determine whether the researcher should dismiss the null hypothesis or approve the substitute hypothesis.

People experiment with groups to discover if there is a distinction among them. The number of independent variables (IVs) in an ANOVA test is referred to as one-way or two-way.

  • There is only one independent variable in a one-way analysis (with 2 levels).
  • There are two independent variables in a two-way analysis (it can have multiple levels).

Figures 2 and 4  show the average participation of consumers by each purchase in Swedish currency units (SEK).

Figure 2 (Table II in the original research report) shows descriptive statistics for each country’s participation per order.

Figure 2: Contributions from each order by customers

To investigate the variance in input per order in further depth, two-way ANOVAs were performed and the results for all 4 of these regions. The ANOVA for the Swedish sample group is shown in Figure 3 (Table III in the original paper) (the important trends are again similar for all of the regions that are analyzed).

Figure 3: ANOVA results

Figure 4 (Table IV in the original research article) shows descriptive statistics for all 4 nations studied in terms of overall contribution by the customer and the year.

Figure 4: Total number of contributions

The ANOVA for the Swedish subset is shown in Table V of the original article and shown in Figure 5 here. 

Figure 5: ANOVA results and the contributions of customers

Findings gathered by the authors

The e-commerce industry accumulates a massive volume of information, yet this knowledge is rarely exploited to generate service distinction. This research examined forms of behaviour that customers exhibit and found that segmenting customers based on sales and return trends can help with the unique conveyance of desired services.

The relevance of recognizing the dominant purchasing behaviours in a supply chain is highlighted in this research conducted by the authors. The focus of this research was to see if the “one size fits all” method in clothing e-commerce leads to uniform conduct. Although the categorization of clients (Figure 1) in this work is not categorization, it does reveal a varied purchase behaviour and so warrants more investigation into diversified delivery of services.

The researchers note that there is convincing evidence for both assumptions after summarising the study results and relating the outcomes to the overall hypothesis and study objective. Consumers act in a varied manner, as demonstrated by the model theory obtained from the study, indicating that the “one size fits all” idea is archaic, as evidenced by the available literature (Hjort et al., 2013).

The requirement dynamics of the various products/markets provided by a business should be carefully analyzed before deciding on a supply chain system (Christopher et al., 2006, 277-287).

Customer returns are generally regarded as a bad part of conducting a business-related operation; yet, the researchers of this paper discovered that one of the most valuable customers is a regular client who returns things frequently.

The role of the findings for future work

The results of this study show how customers are interacting and show that there is certainly a varied reaction to the “one size fits all” policy from consumers. It is vital to note, however, that classification is merely the beginning of harmonizing the company’s and supply chain’s capabilities.

Qualitative data that provides a solid knowledge as to why consumers behave uniquely should be included in articles published by the subsequent researchers, as it is critical to study value propositions. From the supply chain viewpoint, the authors of this research paper suggest that future studies should investigate techniques for building and providing matched propositions of value.

Practical implications of the research conducted

The results corroborate a diversified service delivery method that includes a somewhat more flexible approach to waste prevention and links supply chain and management initiatives to consumers’ purchasing and returning habits to prevent both of the following conditions:

  • Providing service to the customer beyond the optimal point of customer delight
  • Providing service to customers below the optimal point of customer satisfaction

Humans are all born with a unique set of beliefs, and as expected, each and every one of them will have unique standards when it comes to offerings. As a result, brand classifications and purchasing conduct interplay, but it is purchasing patterns that dictate consumption patterns and, as a result, how we must construct our supply chains in both forward and reverse (RM) orientations (Gattorna, 2010).

Furthermore, the amount of supply chains is ultimately determined by the variety of purchasing behaviour, albeit some estimates may be needed to make this strategy effective.

To Conclude

The significance of RM in service encounters in clothing e-commerce is supported by this study since a big proportion of consumers is consistently returning products. Businesses that use a “one size fits all” strategy, on the other hand, are concentrating primarily on RM performance, losing out on the potential to achieve a competitive edge. 

They are overlooking the benefit that difference could bring to their company, its consumers, and its supply chain partners. A diversified return service could help attract consumers and greatly serve consumers with varying returning habits; this subject is referred to as RM in this research paper.

The amount of supply chains is ultimately determined by the spectrum of purchasing habits, albeit some approximations may be obliged to perform to make this strategy viable.


  • Beck, P., Hofmann, E., & Stölzle, W. (2012, January). One size does not fit all: an approach for the differentiated supply chain management. International Journal of Services Sciences, 4, 213 – 239. 10.1504/IJSSCI.2012.051059
  • Christopher, M., Peck, H., & Towill, D. (2006, May). A Taxonomy for Selecting Global Supply Chain Strategies. The International Journal of Logistics Management, 17, 277-287. 10.1108/09574090610689998
  • Gattorna, J. (2010). Dynamic Supply Chains: Delivering Value Through People (J. Gattorna, Ed.). Pearson Financial Times.
  • Hjort, K., Lantz, B., Ericsson, D., & Gattorna, J. (2013, November 4). Customer segmentation based on buying and returning behaviour. Emerald Insight.