When Data Quality Meets Retail: An Insight
In his book ‘Thinking Fast and Slow’, Daniel Kahneman introduced the concept of ‘Availability Heuristic’. ‘Availability Heuristic’ is a mental shortcut relying on immediate examples which come to a person’s mind while evaluating a specific topic, method, concept or decision. For example, data collection may discover that respondents spend time looking at a website’s blog and eventually the blog helps convert a sale or returning customer.
Obtaining data is difficult, and to pin it to accuracy is even more. Any misrepresentation could lead to bankruptcy. Over the years, retail has depended on data quality for accurate decision-making, determining product launch, buying patterns, merchandising, pricing, marketing and promotions, operations, and customer service support.
With the e-commerce boom, data has become pivotal in determining ways to improve productivity, efficiency, cost-effectiveness in the shipping, delivery process and even to manage customer complaints. However, data dependency has its own set of challenges. Erroneous source of data or insufficient mining can undermine the product lines chosen by the retailers, pricing and promotion of products, and consumer buying behavior.
Ensuring product data quality is pertinent
Being a leading omnichannel, home improvement retailer and a Fortune 50 Company, Lowe’s is driven by data. One of the ways in which it uses data, is to recommend products to shoppers. Having accurate item data improves product recommendations, inventory management, prevents failed or delayed deliveries, and ensures accurate product reviews.
Data helps in making strategic and tactical decisions and eliminates defects from retail operations. It, thus unveils opportunities to focus on the product data quality. This niche, in turn, serves as a ‘competitive differentiator’ in the consumer-oriented retail sector. Data quality also ensures accurate product reviews. Since online retail stores lack physical access to products, accurate and genuine customer reviews help customers in decision making.
High data quality is essential in retail especially pertaining to product information and images. Incorrect data or lack of information can cost heavily to the business resulting in a sharp decline in sales. This may damage business-vendor relationships and consequently, dissatisfied vendors may switch to competitors.
Now, the errors could be related to incorrect product information or data conflict. If not corrected sooner, these could lead to a loss of customer loyalty and vendor security. Vendors are responsible for providing product related data and information. While doing so, they must ensure that the data is in line with standards laid down across multiple retail channels. When vendors don’t adhere to these standards, errors arise. This is when the Data Quality team steps in.
Drivers of Data Quality team
To focus on building competitive ‘data as a driving edge’, there’s a dedicated Data Quality team driven by the vision of “Empowering data as an asset for the enterprise by delivering item-health assurance to internal and external customers”. They are also working to ensure consistent quality of data and content across all item portfolios. The data quality team works on the foundations of data completeness, accuracy, quality, systematic data validation, and management of corrected bad data. The primary objectives of data quality teams are
Eliminating Product Data Quality errors: To reduce product discrepancies for a smooth and seamless consumer buying experience and convert prospects into loyal customers.
Maintaining Item Health: The overall accuracy, completeness, and consistency of an item depicts its health across online channels.
Processes in place
Good product data is the pre-requisite for creating a great customer experience. It’s much more than just accurate specifications. The underlying question is: How can you create consistent, relevant, up-to-date, and accessible product information to take the product experience to another level?
This is where a perfect blend of people, processes and automation comes into play. The Data Quality team formulates the best practices, activities, and process automation to solve day-to-day and long-term data quality issues. This is accompanied with building an automated infrastructure to reduce manual interventions i.e., the ‘Data Quality Engine.’
Customer feedback data
Customer feedback collected from product pages is regularly reviewed to accurately capture customer issues related to the product data displayed online. Issues which occur frequently are identified and commented on, raising the questions of what the issue is and where it is seen. These issues can be related to product dimension, images, specifications, etc., all of which are crucial elements that dictate a customer’s buying journey.
Reviewing customer feedback is crucial for maintaining quality criteria for vendors, which in turn solidifies vendors’ trust and customer loyalty. These issues are then enlisted and shared with the vendors to provide updated information and maintain product data quality.
Collection of high-quality data is challenging. Data quality issues may arise when enterprises integrate data systems of different departments, adopt new software, or manage data manually. However, businesses can take certain steps to enhance data quality. Data has become crucial for businesses as data management technology and approaches improve. More businesses are using data to make marketing, product development, financial and other decisions. Higher quality data is empowering enterprises to stay at the top of their game. Those yet to explore the benefits of data risk losing out on future opportunities.
About the author:
Kavya Ganapathy is Manager — Data Quality, Item Integrity Ops & Item Congruency in Lowe’s India. She is a result oriented professional with over 10 years of Retail experience in managing, improving and standardizing operations across multiple geographies including UK, Central Europe & US.
