The company runs a large online auction platform and shopping website which 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
The company earns its revenue primarily through fees collected on its marketplace sales, which includes payment processing and first-party advertising charges. Most of the company’s revenue is earned from the Take Rate on the Gross Merchandise Volume (GMV) of transactions processed on the marketplace. The GMV is the total value of all paid transactions between users on the marketplace platform during the applicable period, inclusive of shipping fees and taxes. The company defines Take Rate as the net revenue divided by the GMV. This Take Rate is a key
performance indicator which allows the company to measure how it is able to monetize volume through marketplace services on its platforms, for a given time frame.
Overall, the sellers on the marketplace are charged on the company’s usage-based fee structure. These fees include the following,
- Final value fee: This is recognized when an item is sold by a seller on the platform. The seller is charged a fixed percentage of the value of each order generated.
- Online store subscription fee: The sellers that want to have a dedicated online store are charged this fee depending on the type of store which is chosen.
- Feature fee: The seller pays this fee to use extra features such as additional text, photos, videos, etc.
- Promoted listing fee: The seller pays this fee for promoted listings of their products on the platform.
- Listing fee: The seller pays this fee for listing of their products on the platform.
In order to significantly boost their revenue growth, the company aimed to revamp the existing price planning program, while also ensuring that the new model did not unsettle their sellers or disrupt existing sales.
THE SOLUTION
The team of business analysts and data scientists from Prescience Decision Solutions selected a set of sellers from the complete list of active sellers on the marketplace platform. These sellers were classified into different segments and analysed according to their country, time of the year, competitor pricing and overall seller characteristics.
Based on these details, our team used various modelling techniques to determine the elasticity of
the fees charged by the company and the possibilities of churn among various categories of
sellers. The different models that were used to build the solution included,
- Dynamic Category Price Input tool which determines the incremental revenue that can be generated by increasing X% of fee s. This was built on Microsoft Excel.
- Price elasticity of Supply and Demand were used to understand the change in a seller’s supply i.e. their listings on the website, in response to a change in platform fees. Our team built a mathematical model which fit the supply curves at aggregated product SKU levels, such as item price range, category, and item condition. With this model, the team compared the behaviour of the sellers affected by the fee changes and compared it to that of the unaffected sellers.
In new markets, where these models could not be successfully applied, different promotions were used to study the behaviour of various seller segments and identify their churn and corresponding price elasticities.
A Price Change Monitoring dashboard was built on Tableau to monitor and report post price change behaviour with respect to forecasts.
This overall solution was applicable to all categories of products, sets of sellers, markets (countries) or for the whole marketplace platform. Each segment of sellers was scored based on their probability of churn and price elasticity. These scores helped the company to identify the seller segments that would readily accept a new fee based model, and which other sellers would reduce or stop listing their products on the marketplace, if the company changed its existing pricing structure. With this, the company was able to accordingly roll out a flexible usage-based fee structure that was customized to various seller segments.
The team from Prescience also enabled extensive automation to reduce the traditional manual steps involved in the pricing related activities. This automation was highly efficient, error-free, data-driven and aligned to market dynamics.
The different technologies used for this engagement included,
1. SQL
2. Microsoft Excel
3. Tableau
THE IMPACT
With this new advanced price planning solution in place, the company had end-to-end control of the flexible price planning process. The solution enabled the company to roll out an optimum fee structure for its sellers, which could be periodically revised, thereby, improving the overall sales on the marketplace and the corresponding company revenues from the take rate.
