Forecasting Buying Intention Through Artificial Neural Network: An Algorithmic Solution on Direct-to-Consumer Brands

 From: FIIB Business Review

In recent years, the organizations are soaring through a more dynamic, uncertain, complex and ambiguous economic environment where technological, environmental and socio-economical changes repetitively challenge the existing competitive advantage making it transient and impermanent. Mounting vulnerability drives firms to device procedure, strategies and operational devices to use customer analytics and other business analytic technologies, a competitive advantage that can explore consumer behaviour patterns through the application of big data analytics (BDA) and data mining.

An increasing number of firms including E-commerce organizations have been adapting BDA in order to predict consumer behaviour which will fructify to better mechanism to create, deliver and capture business value, while strengthening IT infrastructure and delivering operational, managerial, strategic and organizational benefits that can ultimately be translated into a competitive advantage and better performance.

In other words, today’s business environment is hypercompetitive and intricate with multiple economic factors which compete, cooperate or even cooperate to control and analyse enormous volumes of data.

So far, most management scholars have researched value creation by understanding the online buying behaviour of the consumers through the deep study on their perception, attitudes, pricing decisions and clustering based on their online buying behaviour. Accordingly, the extant literature has focussed on several areas like online buying behaviour, service quality of the online shopping portals, and digital marketing aspects of the E-commerce organizations. However, very few of them have empirically addressed the issue of segmenting the consumer on the basis of their attributes of online buying behaviour through the Deep Learning methods and artificial neural network (ANN) model. The top-performing firms use data more as compared to the lowest-performing firms. This area is particularly important for direct-to-consumer (DTC) brands operating in fast-changing and dynamic markets, as these firms must apply the algorithm-based models of deep learning and ANN to maintain their competitive edge.

Despite scattered anecdotal evidence about the practical use of deep learning, big data analytics and the emerging literature on big data, deep learning methods, it appears that there is a knowledge gap spinning around how DTC brands are using deep learning methods to understand their customers, which will further lead to an efficient business model.

The main objective of this study is to bridge the aforementioned knowledge gap by extracting an ANN model based on the buying attributes of the consumers of the DTC brands. As such, the study aims to address the following research question ‘how a predictive model based on the ANN can be extracted which will help to classify consumers based on their buying attributes’. This question has been addressed by extracting an ANN model. The classification of the consumers based on the ANN will help the DTC brands to understand the buying attributes of consumers.