Source: bloncampus.thehindubusinessline.com
The Covid-19 outbreak has been one of the worst tragedies to befall mankind. Its impact is felt by almost all industries and businesses around the world. The prolonged lockdowns, due to the debilitating effects of the pandemic, have led to a huge decline in demand for goods and services, which in turn, is causing revenues to nosedive.
This unprecedented health crisis is further creating a negative impact on the already low consumer sentiments within the global retail sector. India is the second largest employer providing direct or indirect employment for about 50 million people with a monthly business volume of about $75 billion.
While the debate continues on whether Covid-19 is a black swan or the new normal, businesses have to orient their models to suit the ‘new normal’ ways of the customer. That would mean bringing in a few dimensions that could help project the big picture better and working on them to mitigate the risk, increase market share and thereby mobilising the wheels of economy.
Today, due to the social distancing norms, customers who would otherwise prefer visiting a physical retail outlet may have to embrace e-commerce platforms for fulfilling their needs. Given the above reality, the businesses that understand their customers better and engage with them in meaningful ways, can retain them. Catering to the customers’ needs through alternate channels is the only way to mitigate losses caused due to the pandemic. Data science, or data collection, integration, governance, and analysis can be leveraged to navigate successfully through this crisis.
Seven key areas where Data science can helps retailers are:
Identifying shopping pattern
Data analytics helps in understanding the customers better through their buying patterns. For instance, it captures all the interactions the customers have with the brand; what kind of products they view, in what price range,what time of the day, week, or year they log in to the online store, what kind of products they add to their cart, their click through rates, and so on.
Cross-selling and upselling
Affinity analysis models such as market basket analysis, one of the key techniques used by large retailers, is a popular machine learning technique involving a set of statistical affinity calculations. It is used to study combination of products that most frequently occur together in orders and uncover the association between these various products that customers buy. These relationships can be used for cross-selling and can be further explored by other data science tools to curate product promotions, upselling techniques and better recommendations for the customers. Large retailers, such as Amazon and Netflix, use this extensively.
Data-backed price management
Today, customers have a wide range of choices for products at the lowest rates. They quickly navigate through different sites, compare the prices and choose good deals. Big brands like Amazon, use data analytics to process large volumes of data, to process competitor’s prices, product sales, customers actions and geographical preferences for developing dynamic pricing algorithms. Amazon’s software is built to manage consumers’ perception of price. It can identify the goods that loom large in consumers’ perceptions and keep their prices carefully in line with competitors’ prices, if not lower. The prices of all the other products are allowed to drift upward.
Customer classification and segmentation
Predictive analytics classification models such as Discriminant analysis and Logistic regression can classify customers based on certain variables into loyal customers and brand switchers. This information can be used to design various marketing strategies to reward the loyalists and have more meaningful engagement with the brand switchers to minimise the customer churn. Machine learning algorithms such as K-means help market segmentation and predict how likely a customer segment X will respond to a 10 per cent discount on a certain product or how a particular consumer responds to a certain combo offer.
Brand perception and positioning
For any business to stay in the competition, it becomes imperative to understand how the customers perceive their brand. Machine learning algorithms and multidimensional scaling helps answer questions such as what goes on in the customers’ mind when they think of their brand, where do the customers place their brands or products in comparison to their competitors, and mainly who is perceived to be their competitors. They also provide robust spatial maps to understand each dimension better enabling the businesses reach out to customers with more precision and strategise their new product launches and product positioning more effectively.
Supply chain
When customers place orders on an online platform, they expect information on order tracking services while the goods are still in transit. Even well-known brands in e-commerce often face difficulty in meeting these expectations, leaving the customers dissatisfied. This happens because the supply chain is dependent on third parties for services such as warehousing and transportation. Under this scenario, using supply chain analytics, that integrates multiple pieces of information from multiple parties on multiple products, helps in catering to the customers’ expectations and fulfilling their needs on product tracking.
Customer satisfaction
While all the above-mentioned areas contribute to customer satisfaction in one way or the other, the overall experience of a customer with the brand or product or service is what determines the satisfaction level. Complaint handling mechanism is the most crucial factor for enhancing customer satisfaction. Customer grievances can now be addressed on Facebook and Twitter, and social media , where the customer reaches out to the brand with his/her complaint and feels acknowledged to receive a response directly on their Facebook page or Twitter handle. Social media analytics models and Sentiment analysis can be used to understand how happy or unhappy customers are and weigh their expectations from the product.
Conclusion
Now, e-retailing is probably all set to create the biggest revolution in the retail industry as a response to the challenges thrown by the pandemic. Retailers should leverage the digital retail channels, by spending less on physical infrastructure and investing more in the data space. They must not only try to retain, but also broaden their customer base from tier 1 cities and reach out to more customers in tier 2 and tier 3 cities, whose buying potential is perhaps not tapped to the extent it should be. Taking clues from the above mentioned data science tools, businesses must take advantage of such times to sail through this tricky situation and make a positive, long-term impact on the customers’ minds.