In today’s digital world, e-commerce websites are the key source of customers for shopping. Information-rich and attractive websites, more customers will get attracted to such websites. Information-rich websites provide information on Pricing, Deals, Promotions, Discounts, Free Samples, etc. in addition to product information and specifications.
The business team builds data using various sources viz. internally shared data (Sharepoint, GDrive, Common Drive), ERP, CRM, Pricing Engine, etc., and from outside websites, and specialized services.
While pulling data from diff sources and systems, some sources/systems are loosely integrated into each other, leaving room for data inconsistency in promotions, deals, discounts, priority flags, etc.
A Case Study
The following case study narrates challenges encountered in maintaining the website with quality data as well as compliance on pricing data in sync at different places it appears on the website as per local laws.
How does data inconsistency impact website effectiveness & CX?
During website browsing, when the customer observes products displayed with mixed tags e.g. New Arrival section shows items listed with Clearance tags, the Product with Promotion End date in the past, the Product of one OEM mixed with another OEM offers, etc, i.e. conflicting or wrong information make customer nervous. It further adds to the frustration when applying filters on search does not remove such discrepancies. This makes customers turn to other websites for their needs.
Do such data discrepancies also affect on the pricing models?
With multiple data sources, business rules, and multiple teams handling pricing related data before publishing to the website, there is always the risk of data inconsistency, and the possibility of wrong data rendered on the website for Promotions, Deals, Discounts, validity dates, etc. Simple priority flag cause to show the different price on Product List Page Vs Product Data Page Vs Product Price when added to the cart.
Any such discrepancies confuse customers and lead customers to look for alternatives by giving poor feedback and occasionally registering feedback with independent agencies.
In some countries, such price discrepancies are treated as violations and compliance issues, resulting in a penalizing E-Commerce website.
Why testing does not uncover all such issues during testing?
The testing team does carry out testing with multiple regression cycles to uncover all discrepancies by performing multiple testing cycles and reporting defects using test data provided by the product owner. The test data is limited by data volume and data quality, as most of the time test data is created manually, which limits testing to specific scenarios leaving edge cases open.
Additionally, the testing team looks for the assertion of specific scenarios & outcomes for success or failure as per business rules and does not necessarily test for logical errors i.e. Conflicting discounts, items mixed across OEMs, Free Sample availability, etc.
The product team performing manual validation, should not cover these risks?
Performing such validation manually will be limited to a small data set and may not cover all edge cases. Additionally, large efforts or longer cycle times are required to perform these validation affecting the goto market schedule viz. release timelines, release features, promotions, and discount plans.
Removing Erroneous flags and promo end date:
A Software Bot built to follow the manual steps of the product team to browse thru the website New Arrivals Menu > Sub Menu and browse all items listed on multiple product list pages. Software Bot trained to check erroneous ‘tags’ and promotion end dates in the past for all items. Information of such items with inaccurate items was made available to the product team, which in turn worked with Development and other teams to sanitize data and improve validation rules before publishing data to the website.
Software Bot run frequently to keep eye on any data appearing on the website with erroneous tags or promotion end date in the past and provide information to the product team for corrective actions.
Key Benefits: Improved data quality and valid promo dates along with improved customer experience. Saving 80%+ manual efforts and timely release of promotions on the website.
Identifying price discrepancies on Product List x Product Data x Product into the Shopping cart
This activity was performed using software Bot integration with REST API endpoints used on the website, which bring data from different sources and render it to the website page.
Software Bot built to fetch all items using search API (option to read from web pages) and evaluate items with promo and/or discount flags. Loop thru every item to fetch info from respective REST APIs viz. Product List, Product Data, and Item addition to the shopping cart. Prices appearing at these three levels are compared and discrepancies are reported back to the product team to fix the issues.
Key Benefits: Consistent price appearing on different pages, compliance ensured. Website sanity checking Bot execution at multiple frequencies, ensured these discrepancies are removed at the earliest. 100% data validation and no edge cases left to guess.
Software bots can be used beyond business process automation, i.e. to ensure compliance and enhance the customer experience of E-commerce websites and save organizations from legal issues and penalties.
Software bots can fill the need for 100% data validation and ensure compliance with website data and improve customer experience.