Flagship Research Quarterly of the Indian School of Business (ISB). Find out more

Sign up  |  Log in

Past Issue • Apr-Jun 2012

Do All Product-Oriented Web Technologies Reduce Product Returns?

Retailers are increasingly using product-oriented web technologies to minimise product returns. Are these technologies effective and how do they influence returns? Professors Prabuddha De, Jeffrey Hu and Mohammad S Rahman present the results of their study.

Product returns are a significant problem for most retailers. Returns cost US manufacturers and retailers almost US$100 billion per year because of product depreciation, and reverse logistics and return rates could be as high as 25% for some retailers.2

The problem of returns is more severe for online retailers as consumers cannot “touch-and-feel” the product before purchasing it in this case. Not surprisingly, retailers – in particular, online retailers – are constantly seeking ways to minimise returns. In recent years, many such retailers have made significant investments on product-oriented web technologies, such as zoom, alternative photos, and colour swatch, to help mitigate this problem. In short, zoom allows a consumer to inspect the finer details of the focal product; alternative photos allow the consumer to look at the product, or a model wearing the product, from different angles; and colour swatch enables the consumer to change the colour of the product to other available colours for better visualisation.

Naturally, it is important for managers to understand how consumers use these product- oriented technologies and how the use of these technologies influences product returns. The normal expectation would be a positive effect, i.e., each of these technologies would reduce returns (and/ or increase sales). But is it really so? Based on a recent study, which we describe below, it turns out that different technologies actually have different effects on returns. In particular, a higher use of the zoom technology leads to fewer returns, as one would expect. However, contrary to normal expectations, we find that the use of alternative photos increases the likelihood of returns. Perhaps more importantly, its use has a negative effect even on net sales (i.e., sales minus returns). In other words, this technology is detrimental to the retailer in whichever way we look at its impact! Colour swatch, on the other hand, does not seem to have any impact on returns.

Contrary to normal expectations, we find that the use of alternative photos increases the likelihood of returns. Perhaps more importantly, its use has a negative effect even on net sales.

Study Description and Setting

The research setting of our study is the retailing of clothing products on the Internet, an industr y that is currently undergoing rapid adoption of advanced product-oriented technologies. We have obtained data from a women’s clothing retailer that has implemented all of the product-oriented technologies mentioned above, namely, zoom, alternative photos and colour swatch. The company’s website provides consumers with navigational features such as browsing, a search function and a recommendation system, which help them access a broad set of products. Once they are on a product page, they can use any of the three product- oriented technologies available. Matching each consumer’s technology usage with her returns requires the identification of her website sessions. Fortunately, when a consumer makes a purchase online, the same web order identification is recorded in both the ser ver log and the purchase database, enabling us to identify all her purchase sessions. Thus, by combining the ser ver log with transaction data, which also records whether a purchased product was returned or not, we are able to measure the use of each of the technologies by a consumer for a particular product by counting the number of times the focal technology was used for that product.

Our study shows that, contrary to normal expectations, all product-oriented technologies are not equally good for the retailer. In fact, the alternative photo technology is always found to be detrimental to the retailer, both in terms of returns and net sales.

We use robust econometric models to study the impact of using a specific technology on the likelihood of return. In addition, we account for the impact of technology usage on sales in a two-stage model, where the first stage considers the impact of each technology on sales and the second stage captures the technology’s impact on returns. Thus, we are able to quantify the effect of technology usage on net sales as well. This is the first empirical study that examines product-oriented technologies closely and quantifies their impact on returns and sales.

Opening the Black Box

The key to deciphering the impact of product-oriented technologies on returns (in particular, why zoom usage leads to lower returns, while alternative photo usage results in higher returns) is to examine the fundamental mechanisms behind returns. As shown in prior literature, the primar y determinant of return is dissatisfaction with the purchased product. In general, the confirmation or disconfirmation of pre-purchase expectations drives consumer satisfaction or dissatisfaction.3 Specifically, when the perceived quality (or performance) after a consumer receives a product matches her pre- purchase expectation, her expectation is confirmed and she becomes satisfied with the purchase. On the contrar y, if the post-purchase perceived quality turns out to be lower than the expectation, it disconfirms the expectation, which, in turn, leads to dissatisfaction. Thus, the primar y stimulant of product returns – dissatisfaction – represents the degree of disparity between the expectation and the perceived product quality.4

An important point to note here is that the pre- purchase expectation itself can influence the perceived product quality. If the gap between the expectation and the perceived product quality is small enough to fall within a consumer’s “latitude of acceptance,” then the perceived quality often moves closer to the expectation and the so-called “assimilation” effect sets in. In contrast, if the gap extends beyond the latitude of acceptance, then the consumer tends to amplify the difference between the expectation and the perceived product quality, i.e., the perceived quality diverges further from the expectation, which is characterised as the “contrast” effect.4

Given the discussion above, the breakthrough to the puzzle concerning returns lies in understanding how pre-purchase expectations are shaped. Previous studies have established that there are essentially two types of product information – factual information and impression-based (or evaluative) information.5

In short, factual information refers to “logical, objectively verifiable descriptions of tangible product features,” whereas impression-based (or evaluative) information conveys “emotional, subjective, impressions of intangible aspects of the product.” Factual information leads to a more realistic pre-purchase expectation, which results in a smaller gap between the pre-purchase expectation and the post-purchase perceived quality and facilitates the assimilation effect to set in, thereby reducing the chances of return.

The last piece of the puzzle is to identify the type of information one gains from each of the product-oriented technologies. Following conventional techniques, we have used independent raters to identify the dominant type of information provided by each technology for each product, and found that the information zoom provides is primarily factual, whereas the information alternative photos provide is primarily impression based. The information provided by colour swatch is found to be neither predominantly factual nor predominantly impression based. What are the reasons behind these findings? With the use of zoom, consumers are able to see the finer details of the product – its fabric, pattern, print, stitches and small decorative features – which are all factual in nature. By using alternative photos, on the other hand, they see pictures of a beautiful model wearing the focal product, typically against a scenic backdrop, from different angles. These pictures, taken by professional photographers, are designed to convey ideas or impressions regarding how a consumer may look herself while wearing the product. Hence, while the consumer may gain some factual information by obser ving the model from different angles, the use of alternative photos is likely to draw her attention more toward impression-based information. Moreover, it is known that one’s impression is often guided by the most positive image one sees6, thereby inflating the pre-purchase expectation further. Finally, in our context, by using the colour swatch technology, a consumer can see the product in each of the available colours separately on a larger frame and may obtain a more vivid view of the colour than just by looking at the small colour patch shown on the main product page. While a more vivid viewing of the colour helps the consumer gain factual information, this technology also enables her to visualise how the focal product in a certain colour would look on a beautiful model in a scenic environment, conveying impression-based information to her. Thus, the net effect of colour swatch usage is that neither type of information dominates.

In sum, while studying the impact of a specific technology on returns, it is important to determine the dominant type of information the technology provides. Given the dominance of factual and impression-based information in what one gains from zoom and alternative photos, respectively, it is not surprising after all that zoom usage reduces returns, whereas alternative photo usage increases returns.

The need to carefully consider the type of information while examining product returns is a major take-away from this study

Additional Insights

Do our results remain the same across different categories of clothing? The products of the retailer in question can be divided primarily into two categories – swimwear and fashion clothing. When we consider these two categories separately, we observe that the results show a stronger impact of technology usage in fashion clothing compared to swimwear. This contrast is in line with the fact that there is relatively less room for gaining information to influence pre-purchase expectations in swimwear compared to fashion clothing. This is because the items in the swimwear category are quite standardised, whereas those in fashion clothing may vary greatly. Consequently, in the case of the latter, consumers may have a stronger inclination to form pre-purchase expectations. Moreover, they now have more room to glean information – either factual or impression based depending on the technology used – to form these expectations.

Do the results vary depending on whether the consumer is loyal to the retailer in question (or not)? Interestingly, we find that loyal consumers are more likely to return compared to those who are not loyal after using the alternative photo technology. This is perhaps due to the fact that, because of their familiarity with this retailer, loyal consumers may have more confidence in its products and they may also be more confident about their own quality evaluation of these products.3

Therefore, for the same gap between the pre-purchase expectation and the post-purchase perceived quality, a loyal consumer may be more disappointed or frustrated than a non-loyal consumer.7 As a result, the contrast effect would be more prevalent in this case, thereby making the gap larger. Consequently, when using alternative photos, loyal consumers would be more prone to return the product compared to non-loyal consumers.

Closing Thoughts

Our study shows that, contrary to normal expectations, all product-oriented technologies are not equally good for the retailer. In fact, the alternative photo technology is always found to be detrimental to the retailer, both in terms of returns and net sales. This study is, however, based on the data collected from a retailer of women’s clothing. Will alternative photos remain detrimental for other product categories as well? Not necessarily. The situation could be quite different for a product category such as laptops. There, pictures from different sides might provide important facts, for example, how many and what type of slots there are to hook up other devices or cables. Thus, alternative photos would in this case provide primarily factual information rather than simply creating impressions. It is, therefore, crucial that we correctly determine which type of information is predominant while analysing the impact of the alternative photo technology, in particular. In fact, the need to carefully consider the type of information while examining product returns is a major take-away from this study.

1. Details of the study can be found at http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2077135.

2. Blanchard, D. 2007. Supply chains also work in reverse. Industry Week, http://www.industryweek.com/ReadArticle. aspx?ArticleID=13947, Cleveland, OH.

3. Cadotte, E.R., R.B. Woodruff, R.L. Jenkins. 1987. Expectations and norms in models of consumer satisfaction. Journal of Marketing Research 24(3) 305-314.

4. Anderson, R.E. 1973. Consumer dissatisfaction: The effect of disconfirmed expectancy on perceived product performance. Journal of Marketing Research 10(1) 38-44.

5. Holbrook, M. B. 1978. Beyond attitude structure: Toward the informational determinants of attitude. Journal of Marketing Research 15(4) 545-556.

6. Chowdhury, R., G.D. Olsen, J.W. Pracejus. 2008. Affective response to image in print advertising: Affect integration in a simultaneous presentation context. Journal of Advertising 37(3) 7-18.

7. Woodruff, R.B., E.R. Cadotte, R.L. Jenkins. 1983. Modeling consumer satisfaction processes using experience-based norms. Journal of Marketing Research 20(3) 296-304.

ABOUT THE AUTHORS

  • Prabuddha-De

    Prabuddha De

    Accenture Professor of Information Technology at the Krannert School of Management, Purdue University.
  • Yu-Jeffrey-Hu-Feb7

    Jeffrey Hu

    Sharon A. and David B. Pearce Professor, Director of China Program, Co-Director of Business Analytics Center, and Associate Director of Master of Science in Analytics, Scheller College of Business, Georgia Institute of Technology.
  • mohd

    Mohammad S. Rahman

    Assistant Professor of Management Information Systems and Fellow, Centre for Digital Economy at Haskayne School of Business, University of Calgary.
Scroll To Top