Dynamic Price Implementation
In this article, I explain how to implement dynamic pricing.
My implementation steps:
1- A/B tests
2- Item price determination
3- Price Simulation
4- Income Simulation
5- Best and 2nd Best Scenario
Revenue increased 8.82% in best-case scenario. Revenue increased 7.04% in the second best scenario.
Dataset Story
A game company gave gift coins to its users for item purchases in a game. Users buy various vehicles for their characters using these virtual coins.The game company did not specify a price for an item and allowed users to buy this item at the price they wanted.
The data set includes the fees paid to the items in 6 different categories.For example, for the item named “shield”, users can buy this shield by paying the amounts they see fit.In other words, a user can pay with 30 units of the virtual money, and the other user with 45 units.Therefore, users can buy this item with the amounts they are willing to pay for themselves.
Variables
- Category id
- Price
Python Implementation
The test results with and without outliers gave very similar result but I will use non-outlier dataframe to avoid misleading results
Conclusion
Best Scenario (but risky)
If we use constant price method for non similar groups and flexieble median price method for similar groups, expected max income:
Real income:
If we use constant price method for non similar groups and flexieble low confidence price method for similar groups, expected income:
