Algorithmic Pricing


“The single most important decision in evaluating a business is its pricing power.”


Warren Buffett


E-Tail Genius was selected to work with an international wholesaler of technical products. The wholesalers sells an extensive assortment of swimming pool filters, pumps, heat pumps, liners and accessories in several countries.

The challenge

Pricing has a direct impact on the profitability and therefore the ability to grow. Most pricing decisions require a trade-off between margin and price perception. To avoid a “race to the bottom”—the self-defeating exercise of trying to beat every competitor’s price on every item—retailers and wholesalers must hone their ability to make smart pricing investments.

The prices at this wholesaler were based on a cost price plus method where customer behavior and profitability per customer have not yet been taken into account. It was likely that the company:

  • could add volume by lowering some prices
  • could raise some prices without losing clients

The wholesaler expected to be able to improve their margin by making more educated decisions on their pricing. The first step is to shift from cost price plus methodology to predictive value based pricing and the second step is to shift to dynamic pricing. 

The Solution 

The wholesaler engaged E-Tail Genius to build a solution that could be used to answer what’s the optimal price for every product per customer group. To build this solution we took two steps:

  • The extensive assortment of the wholesaler is divided in two types of products: key value items and background Items. Key value items (KVI’s) are the items which drives the price/value perception for customers and background items are the items that don’t drive price perception and are commonly bought together with the KVI’s. By lowering the the KVI’s in price, the wholesaler will add sales volume. And by raising the price of the background items, the wholesaler can increase his margin without losing clients.
  • The demand to products (and the optimal price) is dependant of factors such as the different customer groups, seasonality, sales regions (location), competition prices, the size of the basket and the cost price. To include all these factors into the optimal price decision, a Bayesian Network Model has been developed. This algorithm calculated what’s the optimal price for every single situation dependent on the factors.

Often, AI systems make decisions, but end users are in the dark about how conclusions were reached, and if challenged the systems themselves can’t provide any reasoning retrospectively. Our technology provides much-needed clarity to understand how and why a decision was made. The end user can follow the decision making process of how the optimal price is calculated.


The Outcome 

When using Algorithmic Pricing:

  • business and pricing analysts spend less time on finding out what’s the optimal price and have a powerful model that includes all relevant parameters in the decision;
  • marketeers can significantly spend less time on the promotional process by knowing which item will add the most value to the customer;
  • management can control the predicted revenues and margins;
  • the company has an integrated solution where the price change of the platform is connected to the clients’ ERP systems;
  • the company shift from a general cost plus price method to a dynamic value based pricing.

Our ability to explain its own actions is particularly important to the pricing project, because all the different sales regions and employees should understand why a price change is a good decision.

Our solution is expected to increase the margin while reducing the complexity of the pricing decision. 

Find out what value we can add for your company