線上消費者再購行為之預測-以Yahoo!奇摩拍賣女裝上衣為例

Prediction of Online Consumers’ Repurchase Behavior on Women’s Shirt at Yahoo! Auction Website

廖致淵(jr-yuan Liao)

指導教授 ： 何靖遠

國立中央大學/管理學院/資訊管理學系/碩士(2012年)

摘要

獲取一位新顧客的成本是留住一位舊顧客的五倍；顧客再購率提高5%，企業的獲利可以提高25%-95%，企業必須要更專注在現有的顧客，才能以更具成本效益的方式增加獲利。由於網路購物的便利，消費者的選擇亦無時空的限制，若能有效地預測現有顧客的再購行為，無疑地將成為電子商務業者的競爭優勢。
本研究以網頁內容探勘的方式蒐集線上拍賣網站所揭露的買賣雙方交易記錄，分析線上買家的忠誠類型，並建立一個綜合RFM、評價和忠誠類型的再購行為預測模型。本研究以“Yahoo!奇摩拍賣”平台2012年3月份女裝上衣有評價記錄的買家為基礎，蒐集這些線上消費者在2012年4月1日以前的交易資料，據以分析其忠誠的再購類型，以及這些買家在4月1日以後兩個月內的再購情形。邏輯斯迴歸分析結果顯示，最近一筆交易時間的間隔愈短、交易總次數愈高、累積交易金額愈大、平均交易金額愈低、最後一次交易給予的評價愈佳、賣家的累積評價排名愈高，和買家屬於單忠誠類型等，買家在一個月內再購的機率愈大。基於本研究所建立的預測模型可以有效地預測線上消費者的再購行為，對於網路購物平台業者和線上賣家均提供有用的經營策略指引。

關鍵字

忠誠類型 ； 再購行為 ； 交易評價 ； RFM模型 ； 網路購物

並列摘要

“Acquiring a new customer is five/six times more costly than retaining an existing one.” “Increasing customer retention rate by 5% increases profits by 25% to 95%.” Given these statistics, seller has to focus on existing customers to increase profitability in a cost-effective way. Due to the convenience of Internet shopping and unlimited choices for online consumers, accurate prediction of the repurchase behavior of existing customers would undoubtedly become its competitive advantage of e-commerce business.
In this study, we use web content mining technique to collect real transaction records between buyer and seller at online shopping website, analyze the loyalty repurchase type of consumer, and establish a comprehensive repurchase behavior prediction model containing RFM, rating and loyalty type. We collected the buyers who had rating records of women''s shirt category from Taiwan’s Yahoo! Auction website in March 2012, and collected their historical transaction data with all sellers of women’s shirt category before the April 1st of 2012, with which we analyze the loyalty type of these buyers. Then we collected repurchase data in the following two months for those buyer-seller pairs traded before April. Logistic regression analysis showed that the more recent the last transaction, the higher the frequency of past transactions, the higher the cumulative transaction amount, the lower the average transaction amount, the better the rating of the last transaction, the higher the ranking of cumulative rating of the seller, and the buyer is loyalty to one seller, the probability is higher for the buyer to repurchase from the same seller in next one and two months. Given the prediction model can effectively predict the repurchase behavior of online consumers, this model provides useful guidelines of business strategy for online shopping platform providers and online sellers.

並列關鍵字

Internet shopping ； Loyalty type ； Repurchase behavior ； Rating ； RFM model

參考文獻

Anderson, R. E., & Srinivasan, S. S. (2003). E‐satisfaction and e‐loyalty: A contingency framework. Psychology and Marketing, 20(2), 123-138.

Ariely, D., & Norton, M. I. (2008). How actions create-not just reveal-preferences. Trends in cognitive sciences, 12(1), 13-16.

Boulding, W., Kalra, A., Staelin, R., & Zeithaml, V. A. (1993). A dynamic process model of service quality: from expectations to behavioral intentions. Journal of marketing research, 30(1), 7-27.

Buckinx, W., & Van den Poel, D. (2005). Customer base analysis: partial defection of behaviourally loyal clients in a non-contractual FMCG retail setting. European Journal of Operational Research, 164(1), 252-268.

Dholakia, U. M., Basuroy, S., & Soltysinski, K. (2002). Auction or agent (or both)? A study of moderators of the herding bias in digital auctions. International Journal of Research in Marketing, 19(2), 115-130.