Predictive Performance of Customer Lifetime Value Models in E-Commerce and the Use of Non-Financial Data

Abstract

The article contributes to the knowledge of customer lifetime value (CLV) models, where extensive empirical analyses on large datasets from online stores are missing. Based on this knowledge, practitioners can decide about the deployment of a particular model in their business and academics can design or enhance CLV models. The article presents predictive performance of selected CLV models: the extended Pareto/NBD model, the Markov chain model, the vector autoregressive model and the status quo model. Six large datasets of medium and large-sized online stores in the Czech Republic and Slovakia are used for a comparison of the predictive performance of the models. Online stores have annual revenues in the order of tens of millions of euros and more than one million customers. The comparison of CLV models is based on selected evaluation metrics. The results of some of the models which use additional non-financial data on customer behaviour-the Markov chain model and the vector autoregressive model-do not justify the effort which is needed to collect such data. The advantages and disadvantages of the selected CLV models are discussed in the context of their deployment.

Comparative Analysis of Selected Probabilistic Customer Lifetime Value Models in Online Shopping

“…In our previous studies (Jasek et al, 2018(Jasek et al, , 2019, Extended Pareto/NBD model outperformed other selected models in most of the evaluation criteria. Its results are suitable for e-commerce non-contractual business outputting CLV and probability of a customer being active in the next period.…”

“…There is no study addressing all the models entirely. In previous research studies (Jasek, Vrana, Sperkova, Smutny, & Kobulsky, 2018, 2019, the authors focused on representative models of different model families suitable in an e-commerce environment and their evaluation and comparison. The authors also stated that the results of Markov chain model and Vector autoregressive model which use additional non-financial data about customer behaviour (e.g.…”

“…The authors also stated that the results of Markov chain model and Vector autoregressive model which use additional non-financial data about customer behaviour (e.g. traffic source, delivery address, purchase day) do not justify the effort which is needed to collect such data (Jasek et al, 2019) even though e.g. social media is an important component of marketing for online stores (Vojtko, 2014).…”

“…It vindicated the necessity and suitability of the proposed research -see Sections 1 and 2. In the next phase, probabilistic models appropriate for use in e-commerce according to the results of (Jasek et al, 2018(Jasek et al, , 2019 were selected. The Data understanding phase defined data requirements based on the selected models.…”

Artificial Intelligence in Business-to-Customer Fashion Retail: A Literature Review

“…Recommender [201] Regressionmodels [142,174] RepTree [203] review [121,138,188,190,204,248] RFM [207] R-GCN [242] RNN [80] Robotandpressuremeasurements [146] SEM [239] siamesenetwork [46,75,79,156] SSD [35,92] Survey [53,127,143,150,162,181,191,196,210,219,221] Survey:kanomodel [139] SVM [64,87,153,179] SVM.REPTree [172] SVP [222] UCB [233] VAR [207] VGG-IE [65] Viola-Jones [148] word2vec [98,102,167] word2vecSVMperf [180] XGBoost [176] Table 6<...…”