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
To cater to the varied information requests from their business users and the large number of sellers on the platform, the company has a dedicated Ad-hoc request management team.
Due to the complex nature of these internal and external requests, it took the Ad-hoc request management team several hours to turnaround each one. Over a period of time, the company realized that the majority of these information requests were similar in nature and covered a few broad areas. Also, the insights to be provided in the periodic business reports to the management, took several days to create and put together in presentations. These management presentations followed pre-defined formats and were repeatable in nature.
The company needed to develop a comprehensive solution that automated the process of managing ad-hoc queries. It also required a separate solution which automated the generation of presentations with all the required insights for their management teams.
THE SOLUTION
The team from Prescience Decision Solutions analysed the incoming requests from the business users and the platform sellers. After identifying the overall trends in these requests, the team pre-built Python scripts to generate the required results in a tabular format. The members of the Ad-hoc team request management team only had to update the variables or filters in these scripts, corresponding to each new request. This step brought the total effort per information request down to 15 minutes.
To further automate the process, the Prescience team tagged all the requests to one of the major query categories. Based on this, the team developed the Query Generator tool with an easy-to-use front-end interface which allowed users to select the category and add filters for the site, multiple date ranges, sellers, categories, brands, items, condition (new or used) and so on. This script is then run in the company’s enterprise tool and the requested details are immediately displayed. These details can be sent to the user’s email ID and downloaded in Excel format. The tool also allows the company to track the templates being designed and their usage.
After analysing the monthly and quarterly business presentations created for the management teams, our business analysts developed the Insights Generator tool. This tool allowed users to select certain filters and the detailed presentation was automatically created and emailed to one or more selected users.
The different technologies used for this engagement included,
- Python
- Streamlit
THE IMPACT
With the automated Query Generator tool, the company was able to reduce the turnaround time for each information request from several hours to a matter of seconds. The automated Insights Generator tool resulted in the presentations being created with 30 minutes as compared to the earlier manual effort of between 3 – 4 days. The Insights Generator tool has proven to be so powerful that the company held a leadership level conference on the capabilities of the solution, to encourage greater adoption across the organization.
The complete automation of these query and insight generation processes has led to such a significant elimination of effort of the entire Ad-hoc request management team, that the company has reallocated all the members to other core business functions which are of high priority.
