USE CASES APPROACH

Delivery Fees represent a growing portion of both the customer basket value & retailer’s revenue. Hence, this component has a growing impact on customer purchasing decision & the retailer’s P&L.

Delivery Fees Optimisation

Setting the price for Delivery fees is critical for Trading performance, our mission is to make it simple & profitable!

Challenges

    1. Isolated price elasticity: Find out how much 1% delivery fee up/down drives the conversion, revenue & profit impact on a per customer segment basis
    2. Retailers usually expect a relatively “flat” pricing grid: Retailer likes to communicate up-front the Delivery fees structure preventing to leverage the cumulated power of pricing dynamism & deep segmentation
    3. End consumers adapt their purchasing behaviours according to Delivery fees grid structure because they’re looking after an overall basket optimisation

Methedology

  1. We deploy an experimentation method to compute price elasticity and isolate the impact of external factors.  We build a clear correlation between Delivery fees variance & their P&L impact
  2. We build a simulation tool that uses elasticity learnt to simulate the impact of delivery fee price changes on the P&L. This allows us to determine optimal Delivery fee pricing configuration to deliver a specific P&L target
  3. Once the configuration is implemented, we then run regular additional experimentations to update the elasticity and simulation tool

Achievements​

  1. Elasticity Curve and Price Illustration are displayed through an extensive performance visualization framework and each price illustration is constantly updated upon the reinforcement machine learning algorithm.
  2. Thanks to a user-friendly interface of our Web App is possible to observe the detailed impact of each price variance and filter for all the exposed dimensions. Moreover, you can easily export all the necessary data directly into Excel.

Other Use Cases