Pregled bibliografske jedinice broj: 1137288
Advanced Analytics Techniques for Customer Activation and Retention in Online Retail
Advanced Analytics Techniques for Customer Activation and Retention in Online Retail // International Conference on Intelligent Computing & Optimization ICO 2020: Intelligent Computing and Optimization / Pandian Vasant, Ivan Zelinka, Gerhard-Wilhelm Weber (ur.).
Zürich: Springer, 2021. str. 1-15 doi:10.1007/978-3-030-68154-8_62
CROSBI ID: 1137288 Za ispravke kontaktirajte CROSBI podršku putem web obrasca
Naslov
Advanced Analytics Techniques for Customer
Activation and Retention in Online Retail
Autori
Matić, Igor ; Mršić, Leo ; Keppler, Joachim
Vrsta, podvrsta i kategorija rada
Poglavlja u knjigama, ostalo
Knjiga
International Conference on Intelligent Computing & Optimization ICO 2020: Intelligent Computing and Optimization
Urednik/ci
Pandian Vasant, Ivan Zelinka, Gerhard-Wilhelm Weber
Izdavač
Springer
Grad
Zürich
Godina
2021
Raspon stranica
1-15
ISBN
978-3-030-68154-8
Ključne riječi
Online, Retail, Web store, E-commerce, Big data analytics, Machine learning, Churn prediction, Prevention and retention
Sažetak
In an age of ubiquitous, super-fast internet, online orders have been increasing exponentially. This, in turn, significantly increases the customer's options in terms of product range and price, and thus has an impact on the increased competition between companies. It was known that customers are often switching between offers and thus between companies or just stayed dormant. The associated decrease in the average order frequency therefore managing customer churn has a huge profit potential for each online retailer. For online retailers, customer loyalty and regular purchase behaviour is an important part of achieving the sales and margin targets so that maintaining and preserving the customer base. This paper uses the key performance indicators of one big online retail company to examine the current situation in detail and provide methods to reduce the churn. For this purpose, several aspects are used, ranging from the use of tracking software to record customer activities and interests in the online shop itself, to the resulting segmentation into various customer types and the precise calculation of customer lifetime value. These aspects converted to the numerical values are used to train machine learning model with goal to calculate a probable churn score. Additionally, the probability calculation for reordering is used as an input for further marketing activities together with estimation of financial uplift and profit potential.
Izvorni jezik
Engleski
Znanstvena područja
Računarstvo, Informacijske i komunikacijske znanosti
POVEZANOST RADA
Ustanove:
Visoko učilište Algebra, Zagreb