Introduction
With the advent of the internet and consumers being exposed to new technology, it has become harder for marketers to predict ever-evolving consumer behavior. Machine learning has allowed studying buyer behavior and has made it easier to anticipate the buyer’s next move for future purchases. Electronic commerce has made it more complex for consumers to buy products as nowadays there are a vast number of options along with comparisons. Overexposure to digital platforms requires a deeper scrutiny of these digital platforms. Data analytics tools are being used to track the digital footprint of the buyers. Marketers need to infer the relevant information from the big data collection from varied digital platforms. When big data analytics come into the picture to analyze customer behavior, it is referred to as customer analytics. By using customer analytics, marketers can draw valuable information and predict buyer behavior, which may lead to enhanced sales, personalized marketing, optimization of the market, fraud identification, and many more applications. Using big data, companies can predict consumer behavior more accurately than with conventional statistical techniques alone (Balar, Malviya, Prasad, & Gangurde, 2013). Big data is emerging as a new power that can drastically change the way companies analyze customer behavior. There are four V’s of big data: volume, variety, velocity, and the fourth one is emerging as the most significant to the marketers, and that is value, as companies understand there is a huge difference in the online and offline behavior of customers. Most of the industries are trying to get one step closer to big data by seeing the various opportunities it has to offer. Retail has always been an important part of our economy and our lives since in order to fulfill our needs (business and customer), we must buy products and services every day. The retail sector is also important for the economy as it significantly contributes to the national economy. To better understand the needs and wants of customers and to satisfy them, most of the companies are getting help from big data. E-vendors can avail market and managerial transaction cost and time effectiveness by using big data analytics as they can easily track buying behavior of an individual and can convert them into regular customers. Consumer behavior data now can be measured and analyzed even for the customer experiences as it can tell what they need, want, and think.
Review of Literature
The advent of the internet has revolutionized the world, and algorithms are ruling it. These mathematical equations are encountered by each one of us – in recommendations of movies, serials, blogs of your choice, to ad pop ups and social media feeds (Finley, 2014). Big data applications are most commonly used by enterprises not for the benefit of companies and organizations but for personalized and customer understanding and experience, and building a long-term rapport with customers in order to make them loyal, as big data helps in predicting what a customer wants (Chen et al., 2012). Big data analytics has proven to be a very effective tool in improving interactions with customers, which aids in effective development of marketing strategies. With numerous brands being able to connect through more channels with consumers, improving and maintain relationships are essential (Uğur & Türkmen Barutçu, 2017). Many industries are successfully applying big data and reaping the benefits of it in terms of increasing profits and growth rate by understanding the real purchase behavior of customers. Using big data analysis helps enterprises and their customers by providing them a competitive marketing advantage as it promotes customization (Linoff & Berry, 2011). The collected data can be manipulated in different ways to figure out the things customers like, prefer, bought together, and what, when, and how to buy (Tirunillai & Tellis, 2014). Big data analytics helps in predicting and segregating the total customer base, based on their demographic and psychographics features.
Tools of Predictive Analytics in Marketing
Predictive analytics is now widely applied by the marketers to understand the customers and frame marketing strategies accordingly. In marketing it can be used in segmenting the market, forecasting, demand pricing, and better customer satisfaction. The following models are widely used for predictive purposes:
RFM Modeling
Recency: customers who have spent money recently on products or services are more likely than others to spend.
Frequency: customers who have spent money repeatedly are more likely than others to spend.
Monetary: customers who have spent more money at a business are more likely to spend again.
The RFM model is being widely used in retail today to understand customer purchasing patterns and behaviors, such as how frequently they are buying, their recent purchases, and their involvement (heavy/low) with the product.
By calculating customer lifestyle value with the help of RFM, various sales and marketing strategies can be framed out (Khajvand et al., 2011).
Black-Box Model
This model deals with the external stimulus response, that is, environmental stimuli, including economics, technology, and culture and marketing stimuli, which in turn include product, price, and promotion. These stimuli lead customers to make their own buying decisions. In this model the marketing mix acts as a stimulus, and customers respond to these stimuli as they can influence purchasing decisions. The internal factors also influence the purchase decision, for example, values, beliefs, lifestyle, etc. The decision-making process is also a part of the black-box model as customers realize the arousal of a certain need and the fulfillment of that need by purchasing or not purchasing a particular product or service.
Personal Variable Model
Under this model consumers make decisions based on internal factors (beliefs, opinions, values, tradition, and goals). The complex model includes both internal and external factors. This model mainly focuses on the internal stimuli and the influence of that internal stimulus on the purchasing decision. Some contemporary models include Bettman, EBK model, Markow model, etc.
Statistical Analysis and Market Research Tools
For efficient integration of customer behavior data into marketing strategies, different techniques have been used such as multiple regression, conjoint analysis, Hypothesis testing, tests for statistical significance, discriminant analysis, factor analysis, and cluster analysis. These tools help in combining the vast data and provide the most relevant factors, which can be taken into consideration for preparing various marketing strategies and tactics. So that only those factors can be targeted which are crucial; cluster analysis provides such a compilation.
Gaining Insights from Big Data Analytics
Increasing reliance on big data for information has made it challenging for marketers to understand and evaluate marketplaces. Converting data into meaningful insights useful for decision makers is one purpose of marketing research (Hyman & Sierra, 2010).
-
Venn diagram: By combining different segments together, an effort is being made to discover hidden relationships and explore customers that bought different categories of products together. This analysis helps marketers in understanding what types of products are generally bought together, so they can be placed on shelves accordingly. Consumers who are not data literate can benefit from this technique as it presents data in a simplified way.
-
Data prefoliation: Identifying customers who have common features and similar behavior, purchase patterns from the data set to frame efficient marketing strategies. Customers having similar tastes and preferences can be targeted easily as a deeper study of that group is possible and tactics can be framed out accordingly.
-
Time series analysis: Prediction about monthly sales volume or orders placed in the near future can easily be anticipated accurately using time series analysis. Time series analysis provides insight for anticipating costs and sales as it gives a certain pattern over years that would assist in identifying and anticipating consumer behavior. Based on the relevant information, marketers can easily frame out various strategies and tactics.
-
Mapping: Under mapping, colors are used to identify consumer behavior. A map is prepared that is divided based on geography and indicates in different colors which products sell the most. A particular region can be identified for catering to the customers, and specific marketing strategies can be framed on that basis.
-
Basket analysis: Most useful insights are inferred from basket analysis such as products often bought together by customers, and when they buy, which customers are not buying and why? The answers to such questions reinforce the understanding of customers by the marketers as to what products can be put at similar places ensuring greater sales and increased revenues as they frequently will be bought together.
-
Decision tree: When it comes to classification, the decision tree has been used as an important tool. It helps in recommending the right products to customers and even helps in identifying potential customers. Probability of choosing a product can be analyzed by using this technique. Predicting and understanding customer behavior is very complex but with the help of big data analysis tools, marketers can modify their marketing strategies to reach the right audience.
Big Data and Consumer Purchase Decision
To access every “P” of marketing, that is, product, price, place, and promotion (McCarthy, 1960) requires collecting data relevant to these P’s, which are then analyzed by using various statistical tools to gain useful insights for decision-making by the marketers. Traditional methods and big data when combined with decision-making helps companies reduce their product failures (Xu et al., 2016). Understanding of big data for success cannot be ignored as “small data” can not be scaled up to tackle complex marketing problems. Big data can alter a five-step traditional consumer purchase decision (Hofacker et al., 2016) as they get influenced by exposure to the vast information available. From the very first stage of need recognition and to the last stage of post-purchase decision, consumers easily get influenced as meaningful information reduces the need for search and recommendation agents, and filtering tools provide automated evaluation of alternatives as customer preferences can easily be analyzed and accessed by a ranking system. Feedback and ratings provided online influence the purchase decisions. With vast social networks, different consumers participate in the evaluation of post-purchase decisions. Presence and participation of consumers in online/virtual communities has made it even more relevant for markets to understand consumers’ group behavior as well.
Prediction in any area has many benefits as it is a combination of advanced statistical analysis, data mining, real time access, predictive modeling, etc. When it comes to benefits of predictive analytics, the major work revolves around security such as fraud detection and chargebacks, which every retailer wants to tackle as it can reduce the overall fraud that can help withdraw products from assortments that are more prone to fraud occurrence. It is also beneficial for the entire E-commerce industry, backed by the features it offers. Predictive analytics for supply chains also serve and help in understanding the demand for customers effectively. This will include all the forecasting related to the delivery and fulfillment of the orders and their returns, etc.
Analytics – The Road Ahead
As per the recent report released by Statista, the big data market is about to increase rapidly from 23.7 billion (US) dollars in 2016 to 92.2 billion (US) dollars by 2026. Analyzing large data has only become possible with the help of big data analytics. Most of the industries are benefiting from predictive analytics and reaping various benefits in terms of correctly analyzing consumer behavior.
Conclusion
Big data is emerging as a new power, which can change the way companies analyze customer behavior drastically. The Indian retail sector is transforming rapidly propelled by rising household incomes, technology advancements, e-commerce, and increased expectations. New innovative technologies are being used by retailers to provide a seamless and unique shopping experience to the customer. With the advent of the internet and consumers being exposed to new technology, it has become harder for marketers to predict the ever-evolving consumer behavior. Machine learning has allowed studying buyer behavior. Electronic commerce has made it more complex for the consumers to buy products as today the availability of vast data can be compiled easily and of course will help marketers for making better decisions. All the digital activities of the consumers can easily be traced in real time like their browsing history, downloading history, etc., which helps marketers analyze the data collected for understanding the future moves of customers. Big data offers various advantages to the marketers and helps in building brand loyalty. The rise in big data and analytics puts a magnifying glass on the consequences that have arisen from the use of the internet in this digital age.
References
Balar, N. Malviya, Prasad, S. & Gangurde, A. (2013). Forecasting consumer behavior with innovative value proposition for organizations using big data analytics. IEEE International Conference on Computational Intelligence and Computing Research, 1–4. doi: 10.1109/ICCIC.2013.6724280. →
Chen, H., Chiang, R. H., & Storey, V. C. (2012). Business Intelligence and Analytics: From Big Data to Big Impact’, MIS Quarterly, 2012 36 (4):1165–1188. →
Finley, K. (2014). Wanna Build Your Own Google? Visit the App Store for Algorithms’, Wired 2014 Retrieved February 3, 2021, from http://www.wired.com/2014/08/algorithmia/ →
Hofacker, C. F., Malthouse, E. C., & Sultan, F. (2016). Big data and consumer behavior: Imminent opportunities. Journal of Consumer Marketing, 33(2): 89–97. https://doi.org/10.15295/bmij.v5i1.101 →
Hyman, M. R., & Sierra, J. J. (2010). Marketing Research Kit for Dummies. Hoboken: Wiley Publishing. →
Khajvand, M., Zolfaghar, K., Ashoori, S., & Alizadeh, S. (2011). Estimating customer lifetime value based on RFM analysis of customer purchase behavior: Case study. Procedia Computer Science, 3: 57–63. →
Linoff, G. S., & Berry, M. J. (2011). Data Mining Techniques: for Marketing, Sales, and Customer Relationship Management. John Wiley & Sons, NJ, USA. →
McCarthy, E. J. (1960). Basic Marketing: A Managerial Approach. Homewood (Illinois): R. D. Irwin. →
Tirunillai, S., & Tellis, G. J. (2014). Mining Marketing Meaning from Online Chatter: Strategic Brand Analysis of Big Data Using Latent Dirichlet Allocation, Journal of Marketing Research 2014 51 (4):463–479. →
Uğur, N. G., Türkmen Barutçu, M., & Uğur, E. (2017). Determining The Dynamics Of Customer Satisfaction In Natural Gas Sector. Business & Management Studies: An International Journal, 5(1), 115–130. →
Xu, Z., Frankwick, G. L., & Ramirez, E. (2016). Effects of big data analytics and traditional marketing analytics on new product success: A knowledge fusion perspective. Journal of Business Research, 69(5): 1562–1566. →