亞馬遜平台競價廣告投放對企業業績的影響

The Impact of Bid Advertising Placement on Corporate Performance on Amazon

Student thesis: Doctoral Thesis

View graph of relations

Author(s)

Related Research Unit(s)

Detail(s)

Awarding Institution
Supervisors/Advisors
  • Qiying Hu (External person) (External Supervisor)
  • Houmin YAN (Supervisor)
Award date8 Jan 2024

Abstract

跨境電商逐漸成為全球化一個必不可少的環節,而在跨境電商行業中,亞馬遜公司擁有全球最大的跨境電商平臺,每天的活躍用戶高達3億多人。亞馬遜採用的廣告形式是競價搜索廣告,它是一種由賣家對關鍵字進行自主投放從而影響產品排名的廣告形式,是亞馬遜電商平臺最重要的廣告模式。賣家想要提升銷量的最主要方式是進行廣告投放,因此廣告投放是從事跨境電商的企業在市場中生存最重要的工具。本文主要研究廣告投放效率,即廣告投放和績效(收入和利潤)之間的關係。

本文首先介紹跨境電商在全球以及中國經濟中的重要作用,並介紹了亞馬遜平臺中競價搜索廣告的特點,明確了廣告投放在跨境電商企業中不可替代的作用。接著採用古道電商公司在2022年5月至6月的廣告投放和績效資料,構建了基準模型:雙固定效應(個體固定和時間固定)模型。回歸結果展示了每增加1元的廣告投放可以帶來約7.68元的總收入和約1.29元的總利潤。在基準模型的基礎上,本文發現廣告收入(廣告利潤)和產品成本分別具有明顯的仲介效應和調節效應。廣告投放先通過影響廣告收入(廣告利潤)再影響到績效,其中仲介效應占到總效應的70%左右,而不同的產品成本也會調節廣告投放和績效的關係。由於亞馬遜平臺中廣告分為自動廣告和手動廣告兩種,本文也從自動和手動廣告投放資金大小入手,分別進行了回歸分析,發現自動廣告更加適合於產品前期的投放,而手動廣告則是後期產品盈利的保證。

古道公司的產品有四種不同的特徵,因此本文做了四類異質性分析,即按照產品四種不同的特徵分別建立回歸模型。第一類特徵是產品所在的類目,本文按照古道公司經營數量最多的八個類目對產品進行分類,分類後通過對每個類目分別回歸,發現五金的產品績效和廣告投放之間的模型解釋力度最高,而工具類產品最低,說明五金類產品的績效和廣告投放之間關係最為緊密,而工具類則最鬆散。同時本文發現邊際收入和邊際利潤最高的也是五金產品,而最低的是戶外產品,也說明了五金類產品能帶來通過廣告帶來更高的回報。第二類特徵是產品開發和銷售對應的業務小組,即將同一業務小組負責的產品作為一類,根據回歸結果可以直觀的由回歸係數量化出業務小組中包含的開發團隊和銷售團隊的業務能力,以及不同的開發和銷售團隊之間的組合默契水準,由此可以優化企業對團隊的管理。第三類特徵是產品銷售的市場。本文發現美國市場最具有潛力,但是廣告投放競爭也最為激烈,廣告帶來的邊際收入和邊際利潤也較低,而英國市場則剛好與之相反,展示出較高的邊際收入和邊際利潤,但是客戶需求潛力不如美國市場。最後一類特徵是產品的星級。按照產品在亞馬遜平臺上的星級分類後,本文發現對於1星至4星的產品,星級越高通常產品績效越好,而5星產品則需要進一步按照是否為剛進入市場的產品進行劃分:若是進入市場的初創產品,則績效表現較弱;若是成熟產品,則會比其他星級都要表現更好,這也符合實際的經營情況。

基於上述的研究,本文能幫助從事跨境電商的企業在亞馬遜平臺上針對廣告投放和績效之間的關係有更深入的瞭解,並且可以有針對性地對不同品類、不同市場、不同業務小組以及不同星級的產品進行不同策略的廣告投放。
Cross-border e-commerce is gradually becoming an essential part of national trade and even globalization. In the cross-border e-commerce industry, Amazon is the largest cross-border e-commerce platform in the world, with more than 300 million active users per day. Bidding search advertising on Amazon's platform is the unique advertising pattern, and the only way for sellers to boost product sales is to invest in advertising, so that advertising is an indispensable tool for cross-border e-commerce companies to survive in the market. This paper focuses on the relationship between advertising investment and performance (sales and profits).

This paper first introduces the important role of cross-border e-commerce in the Chinese and global economy and some characteristics of bidding search advertising in the Amazon platform, so as to clarify the irreplaceable role of advertising investment in cross-border e-co
merce companies. Then, a double fixed effects (individual fixed and time fixed) model is constructed using the advertising investment and performance data of Gudao E-Commerce Company from June to July in 2022. The regression results demonstrate that each additional RMB 1 of advertising investment can bring about RMB 7.68 of total revenue and about RMB 1.29 of total profit. Based on the baseline model, the paper also finds significant mediating and moderating effects for advertising revenue (profit) and product cost respectively. Advertising investment first affects advertising revenue (profit), and then performance, with the mediating effect accounting for about 70% of the total effect, while different product costs also moderate the relationship between advertising investment and performance. Since advertising in the Amazon platform is divided into automatic and manual advertising, this paper also conducts regression analysis by automatic and manual advertising input, separately. The results find that automatic advertising is more suitable for product pre-investment, while manual advertising is a guarantee for product profitability in the later.

There are four different features of products of Gudao e-commerce company, so this paper makes four types of heterogeneity analysis, that is, building regression models based on four different features of products respectively . Firstly, the products are divided into eight product categories, and regress separately. we finds that the products in the hardware category are the most sensitive to advertising investment, while the tools category was usually the least, Suggests that there is the strongest relationship between performance and advertising investment in the hardware category and the loosest in the tools category; secondly, the products are divided by business teams, i.e., products that are developed and sales under the same business team are classified as one category, and from the regression results, we can intuitively find out the business capabilities of the development and sales groups included in the business teams, as well as the level of tacit understanding between different development. The paper also classifies products by the geographical area, and finds that the US market has the most potential, but also the most competitive advertising, while the UK market is the opposite by US. Finally, the paper categories products by their star rating on Amazon and finds that for products with 1 to 4 stars, the higher the star rating the better the product performance, while 5 star products need to be further classified into starter and mature products, with starter products typically performing weaker and mature products performing the best, which is true to the actual business situation.

Based on the above research, it can help cross-border e-commerce companies to have a deeper understanding of the relationship between advertising investment and performance on the Amazon platform, and to target their advertising investment with different strategies for different categories, different geographies, different business groups and different star ratings.

    Research areas

  • Cross-border E-commerce, Advertising Investment, Performance, Double fixed-effects Model, Regression Analysis, Heterogeneity Analysis