استفاده از بسته بندی در حراج های معکوس آنلاین B2B

In industrial purchasing a request for quotation (RFQ) can consist of a single item, but is most often composed of two or more products and/or services bundled together. While such bundles are used in offline purchase negotiations, their criticality is heightened in online auctions due to their usual short duration and constrained bidding environment. Despite this importance, little systematic discussion or evaluation has taken place concerning bundling practice, especially as it relates to the individual items included in the bundle, the overall bundle composition, and the resulting supply base and ultimate bundle performance. This study investigates these issues by developing a conceptual model and testing it with a large-scale survey completed by purchasing professionals practicing bundling in B2B online auctions. Results indicate that crafting a more homogeneous bundle is most important in achieving a successful outcome. Furthermore, more complex or difficult-to-specify items do not impact perceived bundle performance, but do influence the supply base that is willing and able to bid on the business.

One of the more important components of the request for quote (RFQ) process is the determination of the appropriate order lot (bundle) that a supplier evaluates and quotes upon. The RFQ order lot may consist of a single item, but will more often be comprised of a set or bundle of different items that is attractive to one or more potential bidders. Within this context, bundling (a.k.a., aggregating, lotting and combining) is defined as the aggregation of two or more items (products and/or services) by the buyer into a bundle that is put up for bid to potential suppliers as part of a single RFQ. As such, the composition/configuration of a bundle determines the value of the transaction and the competitive interest between buyers and suppliers, ultimately influencing purchasing performance ( Mabert and Schoenherr, 2001). While bundling has received some academic attention on the sell-side of the supply chain (e.g., Stremersch and Tellis, 2002), few studies on the buy-side have investigated bundling approaches and associated issues. Moreover, while purchasing managers and buyers have performed bundling for centuries, little systematic evaluation has taken place concerning how to guide this important decision, or what variables may come into play and ultimately impact purchase performance. This is surprising, since bundling can be an important element in the general procurement and RFQ development process (Schoenherr and Mabert, 2003 and Schoenherr and Mabert, 2006). Furthermore, while bundling is often employed in offline purchase negotiations, its criticality is heightened in online bidding events (e.g., Beall et al., 2003, Jap, 2002 and Smeltzer and Carr, 2003) due to their usual short duration and constrained environment. Often called reverse auctions, online bidding events are real-time dynamic auctions between a buyer and several suppliers, who compete against each other online, lowering their bid amounts (Beall et al., 2003), until the lowest competitive market price is reached. The constrained environment is created by the fact that once the auction has started, there is usually no possibility for the buyer to modify the bundle, or for the supplier to obtain more information on the quoted items. In contrast, the offline setting provides much more flexibility, and RFQs can easily be modified, even once they have been sent out, to accommodate issues previously neglected or not considered (e.g., clarifications on specifications, supplier suggestions to alter the bundle in order for them to provide a much more competitive quote). This decreased significance may be the reason why bundling has received so little attention in purchasing research in the past. For online bidding events, however, which emerged in the mid-1990s as a new way of doing procurement, all potential concerns must be addressed beforehand due to the inability to modify the bundle configuration during these auctions. If a bundle is not configured correctly, it can lead to a disappointing outcome. This was illustrated in a case study by Mabert and Schoenherr (2001), who present and discuss online auction experiences by a manufacturer of engineered products for the transportation industry. The case study company aggregated 1500 items into 140 bundles for an online bidding event. Although 80% of the bundles got three to four bids on average from interested suppliers, and seven to eight bids on items with a high annual demand and dollar volume, the remaining 20% received unexpectedly either only one or no bid at all. This was a disturbing outcome, since the company is otherwise recognized to be one of the most proactive and advanced firms in regard to sourcing practices. It also illustrated that bundle configuration is a crucial element for success in these settings. Despite this significance, research in the area has been sparse. This paper presents the first focused investigation of bundling for business-to-business (B2B) online reverse auctions. Specifically, this research looks at bundle performance in reverse auctions, i.e. the satisfaction and perceived success of the buyer with the bundle, and identifies performance determinants. These include item specification difficulty, i.e. how challenging it is to specify the items for the bundle, and overall bundle complexity, i.e. the degree of homogeneity or similarity of individual items in the bundle. Moreover, the study identifies how these aspects influence the resulting supply base, i.e. the number of suppliers that are able and willing to bid on the business. In addition, the criticality of this resulting supply base on bundle performance is assessed. To study these issues, a survey of US manufacturing firms using bundles in reverse auctions was conducted. Results, as presented below, provide important practical guidance for operations and supply managers to craft the most appropriate and successful bundle. The paper proceeds as follows. Section 2 reviews relevant literature and outlines the theoretical foundation of this study. In Section 3, the hypotheses are developed and grounded in prior research. Section 4 provides the measure development, information about how the survey was conducted, and a description of the resulting sample. Section 5 tests the model and associated hypotheses with structural equation modeling. Section 6 provides a discussion of the results with detailed managerial implications. Section 7 concludes and outlines avenues for future research.

This study constitutes the first known comprehensive empirical investigation of bundling in B2B online bidding events. It explored the use of bundling in B2B online reverse auctions by focusing on item specification difficulty and bundle complexity, as well as the resulting supply base availability and bundle performance. Research as it relates to bundling in online reverse auctions was reviewed, and a conceptual model was developed based on theoretical foundations in industrial buying, as well as case study insights by the authors. Relationships were tested with structural equation modeling, which received strong statistical support. Analysis of the results provided several key insights regarding bundle composition and its impact on performance. First, item specification difficulty does not directly impact bundle performance, neither negatively nor positively, indicating that highly complex and difficult to specify items can successfully be auctioned. However, an indirect influence of item specification difficulty on bundle performance is possible via supply market availability. Second, item specification difficulty was shown to negatively influence supply base availability, i.e. auctioning highly complex and difficult to specify items will result in fewer suppliers being able to bid on the business. It then must be assessed on a case-by-case basis whether the smaller number of bidding suppliers will be sufficient to achieve the buyer's goals. Third, results suggest that increased bundle complexity leads to decreased bundle performance, highlighting the importance of crafting homogeneous bundles with similar items. Fourth, bundle complexity negatively influenced supply base availability, stressing the need for homogeneous bundle compositions. In addition, while both item specification difficulty and bundle complexity negatively impact supply base availability, the latter has a stronger influence and should be the primary concern when designing bundles. And fifth, supply base availability positively impacted bundle performance, suggesting that the characteristics of a supply base with many qualified and capable suppliers, possessing the necessary capacities, is crucial in online bidding events. While this study led to several important implications for operations and supply managers, there are also limitations. First, the sampling frame chosen could be of concern. The ISM address set provided a convenient sample with which to test the hypotheses and the model. However, the question arises how representative this sample is. Being a member of ISM might already mean that one is more proactive in sourcing activities, simply because one recognizes the value of being a member of this professional association. Therefore, the sample might be biased toward more proactive buyers, and not be representative of the overall industrial purchasing community. And second, the mode of survey administration could be improved. The online environment was the single mode with which the survey was administered at large, i.e., no postal mail, fax or phone contact approaches were pursued with the main sample. Based on the pretest results, the online survey was regarded as the single best avenue to collect data. However, potential respondents may have been lost due to the online administration. They might not have been able to access the survey site due to IT security settings within their company, or they might not have felt comfortable sharing data over the Internet. Some respondents to the main survey indicated the former; they were then sent the questionnaire either as an e-mail attachment, via postal mail, or via fax, depending on their preference. Nevertheless, individuals who did not voice this concern and the willingness to complete the questionnaire if provided in a different mode, were unfortunately lost as potential respondents. Furthermore, the online format restricted individuals to being at the computer when completing the questionnaire. While it is felt that respondents provide more detailed answers to open-ended questions in this environment, due to the ease of responding, further potential respondents might have been lost, who would have preferred to take the survey home with them and complete it at night, on the plane, during the commute with public transportation, etc. In fact, some addressees replied that they were not allowed to complete surveys at work, but that they had forwarded our request to their private home e-mail account, from which they would then answer the questionnaire. Several avenues for future research seem promising. First, the model was tested with data gathered from manufacturing companies. Would the model be validated with samples from other industries? Second, how the use of bundling may be related to other supply management practices, such as spend analysis, strategic sourcing or supply base rationalization, is an important area to address and worthwhile of future efforts. Third, while this study examined bundling from the buyer's perspective, the viewpoint of the supplier has been neglected. For example, how do suppliers react to bundled RFQs and how do they approach their fulfillment? And fourth, analytical extensions could prove beneficial. For example, a decision support model can be developed, based on empirical insights, aiding buyers in bundling the most appropriate products and/or services together. Evaluating if such a decision support approach is useful will shed light on the degree of art and/or science for the bundling process.