線上旅遊網站之再購行為銷售預測

Abstract

隨著資訊與通訊科技（Information and Communication Technology, ICT）的發展及消費者廣泛使用移動設備，使得網路化、行動化的商業應用蓬勃發展，消費者漸漸採用具便利、即時及無地域性的網際網路及行動裝置，生活習慣及消費方式亦即隨之改變。消費者透過社群媒體迅速地蒐集大量資訊，資訊較以往透明化，所進行的交易成本較過往低，對品牌忠誠度亦隨之下降，使各品牌面臨前所未有的挑戰。在這種數位化的型態下，業者必須順應這波潮流積極發展數位化新業務、服務及交易平台。然而，在既有資源下，該如何有效運用資源並找到對的客戶，一直以來都是各品牌的經營要點。 行銷領域中廣泛被研究討論地就是如何利用的顧客購買行為(如一定期間內，顧客交易次數、交易金額、最近一次交易日)等資訊，應用科學方法掌握顧客交易行為的變化，運用不同的機率模型計算出每個顧客的交易頻率及次數，這些預測模型是動態的，預測指標將隨著時間及顧客的交易行為變化而改變。本論文以線上旅遊業為例，並整理過去國外以「機率論」研究消費者線上重複購買行為的行銷科學相關文獻，同時以台灣地區某銀行信用卡於線上消費的資料驗證，並透過Microsoft的Excel來建立銷售預測模型。由於線上消費非屬訂定合約(Non-contractual)型客戶，本研究以機率論研究線上顧客重複消費的可能性，有別於一般較常使用迴歸模型來預測顧客未來交易行為。

With the development of information and communication technology widely used in mobile devices, consumers are gradually using a convenient, real-time and non-territorial Internet and mobile devices. Consumers through social media quickly collect a lot of information, so that transaction costs were lower than in the past and brand loyalty has declined. In this digitized trend, the companies must need to develop the new digital businesses, services and trading platform. The customer relationship has become the important aspect of business management.. In the study, we use a transaction database containing information on the frequency and timing of transactions for online travel shopping customers in order to forecast about future purchasing and through Microsoft Excel to build forecasting model. The forecasting results could develop marketing plans to grasp customer needs. Since most online customers are the non-contractual type, the probability theories study the possibility of online customer repeat purchase, these theories are different from the general regression model to predict future customer transactions.