Enhancing Shipping Efficiency with Analytics and AI-Powered Insights

The company runs a large online auction platform and shopping website that connects buyers and sellers in over 190 countries. Their website has millions of sellers with 1.9 billion global listings and over 132 million active buyers.

THE CHALLENGE

Sellers on the company’s website used different carriers or third-party logistics (3PL) providers to deliver ordered items. The company found that some sellers did not have the required expertise to always identify the most cost-effective shipping options, which added to the landed cost of their products. Also, when these logistics firms encountered operational challenges, it impacted their ability to process pending deliveries from sellers on the platform. These logistics firms did not always provide updated tracking details, which usually led to customers being dissatisfied with the overall buying process. To address these challenges, the company launched a pilot program for managed shipping services in one of its biggest markets. The company tied up with some of the largest logistics providers in the country to ensure the successful shipping, tracking, and delivery of products. With this new managed services program, the company now became responsible for providing live tracking, reverting to customers on the status of their shipped orders, verifying if the product was delivered as expected etc. The adoption rate of this managed services program was expected to be very high. However, in the initial few months, most sellers continued to operate as usual, without opting for this convenient service that was available to them. The company needed an analytics partner to understand why sellers were not utilizing the managed shipping service offering and to recommend functionality upgrades to the program.

THE SOLUTION

The Prescience team found that when the new managed shipping program was launched, most sellers were not aware of this offering, which led to low adoption rates. They also validated whether the core package dimension model was considering the correct default product sizes before calculating the corresponding shipping costs for each product. They found that the available dimension details of some products did not exactly match their manufacturing specifications. As a result, the actual shipping costs would be higher than what the solution originally estimated. This pricing mismatch would have led several sellers to potentially lose confidence in the new managed services shipping program, at an early stage. The team also found that several sellers who had not signed up for the managed services program, consistently failed to revert to buyers if there were any delays in the product shipping. Our analysts recommended that the product team creates a Generative Artificial Intelligence (AI) based interactive customer chatbot which could easily answer common buyer related questions, to help improve customer satisfaction. Some of the key metrics that were tracked included late delivery rate, managed shipping adoption, listing conversions rates, item not received rates, managed shipping vs non-managed shipping performance etc. The team also built a new dashboard for claims related to products that were not as described on their product page, as the company had to cover the costs due to damage that was incurred during shipping. This dashboard gave business users detailed insights into the patterns of claims under this new managed service program. The team also made functionality improvements to the existing set of managed shipping dashboards. To test the effectiveness of the new functionality of the managed shipping program, the team ran an A/B test after dividing sellers into a Test group and a Control group. The different technologies used for this engagement included, 1. SQL 2. Tableau 3. Microsoft Excel

1. SQL
2. Tableau
3. Microsoft Excel

THE IMPACT

With the different analysis, the company was able to identify certain process and technology gaps in the overall managed shipping program and address them. Fixing the data discrepancies in the core package dimension model, led to accurate shipping costs being displayed to all sellers, when they chose the mode of delivery. With the new analysis of the claims records, users were able to get insights into how the company had to reimburse buyers for damaged goods being delivered. Based on the success of this managed shipping program, the company will be rolling out the same service in other large markets in the coming months.